Program, information processing method, and information processing device

Abstract

To provide a program, or the like, for accumulating the contents of conversations of a caregiver regarding conditions of a person to be monitored (care receiver) without increasing the burden on care staff.SOLUTION: A computer acquires sensor data from a sensor which detects condition of a person to be monitored in a care facility or surrounding conditions of the person to be monitored. The computer determines an event which has occurred on or around the person to be monitored on the basis of the acquired sensor data, and outputs the occurrence of the determined event to terminals of a plurality of caregivers. The computer acquires voice data on conversations of the caregivers from the terminals of the caregivers, and converts the acquired voice data to text data. The computer stores the event and the text data converted from the voice data on the conversations of the caregivers regarding the event, in association with each other, on a storage unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present application relates to a program, an information processing method, and an information processing device. [Background technology]

[0002] In a nursing care facility, multiple staff members provide care such as physical care and assistance with daily living to multiple care recipients. In such a situation, for example, each staff member carries an intercom (intercom) and uses the intercom to communicate with each other while providing care, thereby realizing efficient and appropriate care for the care recipients. Patent Document 1 proposes a system that identifies the task target (care recipient) and the task content (content of care work) from the speech of a worker (care staff member), and notifies the worker of an alert if the performance of the identified task does not satisfy preset conditions (conditions for the order in which the tasks are performed, etc.). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-71973 Summary of the Invention [Problem to be solved by the invention]

[0004] In the system disclosed in Patent Document 1, the tasks performed by the care staff can be identified from the voice of the care staff, but the care staff must confirm the condition of the care recipient. Also, in the system disclosed in Patent Document 1, the speech content pattern for performing each task is set, and the care staff must speak in the set pattern, making it difficult to reduce the burden on the care staff.

[0005] The present disclosure has been made in consideration of such circumstances, and its purpose is to provide a program or the like that can accumulate the content of conversations between caregivers regarding the condition of the person being monitored (the person being cared for) without increasing the burden on care staff. [Means for solving the problem]

[0006] A program according to one aspect of the present invention causes a computer to perform the following processes: acquire sensor data from a sensor that detects the condition of a person being monitored within a care facility or the condition of the person being monitored around the person; determine an event that has occurred on the person being monitored or around the person being monitored based on the acquired sensor data; output the occurrence of the determined event to the terminals of multiple caregivers; acquire voice data of the conversation between the multiple caregivers from the terminals of the multiple caregivers; convert the acquired voice data into text data; and associate the determined event with the text data converted from the voice data of the conversation between the multiple caregivers regarding the event and store the text data in a memory unit. [Effects of the Invention]

[0007] In one aspect of the present invention, the condition of the person being monitored is determined, and when an event occurs, the content of the conversation that the caregiver has in response to the event can be stored in association with the event. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of the configuration of an information processing system. [Figure 2] FIG. 2 is a block diagram showing an example of the configuration of a server and a staff terminal. [Figure 3] 10 is an explanatory diagram showing an example of the record layout of a monitored person DB, a sensor data DB, and an event DB. FIG. [Figure 4] 10 is a flowchart illustrating an example of a processing procedure in the information processing system. [Figure 5] 10 is a flowchart illustrating an example of a processing procedure in the information processing system. [Figure 6] FIG. 1 is an explanatory diagram illustrating an example of the configuration of a learning model. [Figure 7] FIG. 10 is an explanatory diagram showing an example of display of conversation text data. [Figure 8] 10 is a flowchart illustrating an example of a processing procedure in an information processing system according to a second embodiment. [Figure 9] 10 is a flowchart illustrating an example of a procedure for creating report information. [Figure 10] FIG. 10 is an explanatory diagram illustrating an example of report information. [Figure 11] FIG. 10 is an explanatory diagram illustrating an example of the configuration of an information processing system according to a fourth embodiment. [Figure 12] 13 is a flowchart showing an example of a processing procedure performed by a server according to the fourth embodiment. [Figure 13] 13 is a flowchart showing an example of a processing procedure in an information processing system according to a fifth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, a program, an information processing method, and an information processing device according to the present disclosure will be described in detail with reference to the drawings illustrating embodiments thereof.

[0010] (Embodiment 1) 1 is an explanatory diagram showing an example of the configuration of an information processing system. In this embodiment, an information processing system is described that monitors the condition of care recipients, such as elderly people and people with physical or mental disabilities, living in each room in a nursing facility, and notifies care staff when it detects a situation in which care from care staff is needed. In the following explanation, the care recipient is referred to as a monitored person, and the occurrence of a situation in which care from care staff is needed is referred to as the occurrence of an event.

[0011] The information processing system of this embodiment includes a server 10, a sensor I / F (Interface) device 20, a staff terminal 30, etc., and the server 10, the sensor I / F device 20, and the staff terminal 30 are communicatively connected via a network N. The network N may be the Internet or a LAN (Local Area Network) provided in a care facility.

[0012] A sensor I / F device 20 is provided for each monitoring target person, for example, and multiple types of sensors provided to detect the state of the monitoring target person and the state around the monitoring target person are connected to the sensor I / F device 20. In the example shown in Fig. 1, a sleep sensor 21, a toilet sensor 22, a door sensor 23, an environmental sensor 24, a human presence sensor 25, and a call button 26 are connected to the sensor I / F device 20 by wire or wirelessly. The sensor I / F device 20 is an information processing device that acquires signals from the sensors 21 to 25 and the call button 26 and transmits the acquired signals to the server 10 via the network N. When any of the sensors 21 to 25 outputs an analog signal, the sensor I / F device 20 is configured to perform A / D conversion, and transmits the signals from the sensors 21 to 25 to the server 10 after A / D conversion.

[0013] The sensor I / F device 20, the sensors 21 to 25, and the call button 26 are provided, for example, in a room (living room) where the person to be monitored lives, but the configuration is not limited to one set of the sensor I / F device 20, the sensors 21 to 25, and the call button 26 provided for one person to be monitored. For example, a configuration may be such that a plurality of sensor I / F devices 20 are provided for one person to be monitored, with the sleep sensor 21 and the toilet sensor 22 connected to separate sensor I / F devices 20. Alternatively, a configuration may be such that a plurality of environmental sensors 24 or a plurality of human presence sensors 25 are provided for one person to be monitored, with the plurality of sensors 24, 25 connected to a single sensor I / F device 20. The sensor I / F device 20 may be appropriately disposed in a position corresponding to the installation locations of the sensors 21 to 25 and the call button 26, or may be built into each of the sensors 21 to 25 and the call button 26. Although only the sensor I / F device 20 is shown for the monitored person B in FIG. 1, the sensors 21 to 25 and the call button 26 are also connected to this sensor I / F device 20.

[0014] One sleep sensor 21 is installed in the bed of each monitoring target. The sleep sensor 21 is, for example, a sheet-like sensor installed between the bed mattress and the sheet to detect the pulse, breathing, body movement, body temperature, etc. of the person lying on the bed. The sleep sensor 21 uses, for example, a condenser microphone, a pressure sensor using the piezo-resistive effect, a pressure sensor using the piezoelectric effect, etc. to detect the air pressure in the air mattress installed on the bed, thereby detecting the monitoring target's entry and exit (getting out of bed), heart rate, breathing state, body movement state while in bed (sleeping), etc. The sleep sensor 21 may also be installed at the head of the bed or on the wall of the room and may use a sensor that uses microwaves or infrared rays to detect the monitoring target's entry and exit of bed and the pulse, breathing, body movement, body temperature, etc. of the person lying on the bed. Based on the detected signals, the sleep sensor 21 analyzes sleep data such as getting in and out of bed, body movements, occurrence of apnea, depth of sleep, occurrence of awakening during sleep (a state in which sleep is interrupted and the person is awake), heart rate, respiratory rate, body temperature, etc., and outputs the data to the sensor I / F device 20.

[0015] Toilet sensor 22 is a sensor installed in a toilet provided in the room of the person being monitored, and detects whether the toilet is in use. Toilet sensor 22 can be, for example, a sensor that detects whether someone is sitting on the toilet (toilet seat) or a human presence sensor that detects the presence of a person in the toilet, and when it detects that the toilet is in use, it outputs a detection signal indicating the detection result (toilet in use) to sensor I / F device 20. Door sensor 23 is a sensor installed on a door for entering and exiting the room of the person being monitored, and detects whether the door is opening or closing, or whether a person has passed through the door. When door sensor 23 detects whether the door is opening or closing, or whether a person has passed through the door, it outputs a detection signal indicating the detected state to sensor I / F device 20.

[0016] The environmental sensor 24 is a sensor that measures environmental data such as temperature and humidity in the room of the person being monitored. The environmental sensor 24 is installed, for example, on a wall surface of the room. The environmental sensor 24 outputs the measured environmental data such as temperature and humidity to the sensor I / F device 20. The human presence sensor 25 is a sensor that detects whether or not a person is present in the room of the person being monitored, and detects the presence of a person by emitting, for example, infrared rays, ultrasonic waves, visible light, etc. and detecting the reflected waves. The human presence sensor 25 is installed, for example, on a wall surface of the room. When the human presence sensor 25 detects the presence of a person, it outputs a detection signal indicating that a person is present in the room to the sensor I / F device 20.

[0017] Call button 26 is a button that is manually operated by the person being monitored when they wish to receive assistance of their own volition, and is installed, for example, on the wall of the person being monitored's room. When call button 26 is operated by the person being monitored, it outputs a detection signal indicating that the button has been operated to sensor I / F device 20.

[0018] In this embodiment, a sensor I / F device 20, sensors 21 to 25, and call button 26 are provided for each monitoring target, and the monitoring target is associated with each of the sensor I / F device 20, sensors 21 to 25, and call button 26. In addition to this configuration, when one sensor 21 to 25 is installed in a location where it may detect multiple monitoring targets, a configuration may be provided in which the monitoring target detected by each sensor 21 to 25 is identified by using face authentication or an ID (IDentification) tag, etc. in combination.

[0019] The sensor I / F device 20 acquires signals from the sensors 21 to 25 and the call button 26, and transmits information corresponding to the acquired signals (information indicating the status of the person being monitored) to the server 10 via the network N. The server 10 is installed, for example, in a nursing care facility and is managed by a nursing care provider that uses the information processing system of this embodiment. The server 10 may also be managed by a provider that provides the information processing system of this embodiment to nursing care facilities. The server 10 is an information processing device capable of various information processing and information transmission and reception, such as a server computer or a personal computer. The server 10 accumulates information about the person being monitored that is sequentially received from the sensor I / F device 20 installed for each person being monitored, and when a situation (event) occurs that requires nursing care, the server 10 notifies the staff terminal 30 of the nursing staff of the occurrence of the event.

[0020] The staff terminal 30 is an information terminal carried by the care staff of the care facility, and is a portable communication terminal such as an intercom, smartphone, or tablet terminal. The staff terminal 30 notifies the care staff of the status (occurring events) of each monitored individual by outputting notification information received from the server 10. The staff terminal 30 is also configured to send and receive voice messages to and from other staff terminals 30, allowing care staff to communicate with each other via the staff terminal 30. In this embodiment, the staff terminal 30 is configured as an intercom with a headset equipped with a microphone 35, earphones 36, and a communication unit 34 (all see FIG. 2 ), but is not limited to this configuration. For example, a headset equipped with a microphone and earphones or headphones may be connected to a smartphone or tablet terminal via a wired or wireless connection and used as the staff terminal 30.

[0021] 2 is a block diagram showing an example of the configuration of the server 10 and the staff terminal 30. The staff terminal 30 has a control unit 31, a main memory unit 32, an auxiliary memory unit 33, a communication unit 34, a microphone 35, earphones 36, etc., and these units are connected via a bus. The control unit 31 includes one or more processors such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), or a GPU (Graphics Processing Unit). The control unit 31 reads a program 33P stored in the auxiliary memory unit 33 into the main memory unit 32 and executes it, thereby performing information processing and control processing that the staff terminal 30 should perform.

[0022] The main memory unit 32 is a temporary storage area such as a static random access memory (SRAM), a dynamic random access memory (DRAM), or a flash memory, and temporarily stores data generated when the control unit 31 executes various programs 33P. The auxiliary memory unit 33 is a non-volatile storage area such as a flash memory, a hard disk, or a solid state drive (SSD), and stores various programs 33P (program products) executed by the control unit 31 and various data. The auxiliary memory unit 33 also stores a conversation application program 33AP (hereinafter referred to as a conversation app 33AP) for transmitting and receiving voice messages with other staff terminals 30 via the network N. The program 33P, the conversation app 33AP, and various data may be written to the auxiliary memory unit 33 during the manufacturing stage of the staff terminal 30, or may be downloaded by the control unit 31 from another device via the communication unit 34 and stored in the auxiliary memory unit 33. The auxiliary memory unit 33 may be another storage device connected to the staff terminal 30.

[0023] The communication unit 34 is a communication module for performing processing related to wireless communication, and transmits and receives information to and from other devices via the network N. The microphone 35 is an audio input unit that collects surrounding sounds and generates digital audio data, and sends the acquired audio data to, for example, the main memory unit 32 for storage. The earphones 36 are audio output units that output audio in accordance with instructions from the control unit 31, and output messages or warning sounds in accordance with instructions from the control unit 31. Note that the staff terminal 30 may be configured to include headphones or speakers instead of the earphones 36.

[0024] In addition to the above-described configuration, the staff terminal 30 may have an input unit that accepts operation inputs from the care staff, a display unit such as a liquid crystal display or an organic EL display, a camera, and the like.

[0025] The server 10 includes a control unit 11, a main memory unit 12, an auxiliary memory unit 13, a communication unit 14, etc., and these units are connected via a bus. The above-mentioned units 11 to 14 of the server 10 have the same configuration as the units 31 to 34 of the staff terminal 30, so a description of their configuration will be omitted. The auxiliary memory unit 13 of the server 10 stores a monitored person DB 13a, a sensor data DB 13b, and an event DB 13c in addition to various programs 13P (program products) executed by the control unit 11. The monitored person DB 13a is a database that stores information about monitored persons in this system. The sensor data DB 13b is a database that stores the status of monitored persons detected by each sensor 21 to 25 and the operation status of the call button 26. The event DB 13c is a database that stores information about events that have occurred for each monitored person. The monitored person DB 13a, the sensor data DB 13b, and the event DB 13c may be stored in other storage devices with which the server 10 can communicate via the network N.

[0026] In this embodiment, the server 10 may be a multi-computer consisting of multiple computers, a virtual machine virtually constructed by software, or a cloud server. In addition to the above-described configuration, the server 10 may also have an input unit that accepts operational inputs, a display unit such as a liquid crystal display or an organic electroluminescence (EL) display, etc. The server 10 may also have a reading unit that reads a non-transitory computer-readable portable storage medium 10a, and may use the reading unit to read the program 13P from the portable storage medium 10a and store it in the auxiliary storage unit 13. The program 13P may be executed on a single computer or on multiple computers interconnected via a network N.

[0027] 3 is an explanatory diagram showing an example of the record layout of the monitored person DB 13a, the sensor data DB 13b, and the event DB 13c. The monitored person DB 13a includes a subject ID column, a name column, an age column, a gender column, a room number column, and a notification target event column. The subject ID column stores an ID uniquely assigned to the monitored person. The name column, the age column, the gender column, and the room number column store the name, age, gender, and room number (room number) of the monitored person in association with the subject ID. The notification target event column stores information (e.g., event name, event identification information) of events that have occurred to the monitored person and that should be notified to the care staff.

[0028] An event is a phenomenon occurring in the state of the monitored person or in the environment surrounding the monitored person, and includes an event that is different from the normal state, an event that should be notified (alert) to care staff, an event that requires care, etc. For example, events include the monitored person getting out of bed (leaving bed), an increase in the monitored person's body movement while in bed (while in bed or asleep), using the toilet and staying in the toilet for a long time, the door being opened or someone passing by, room temperature or humidity outside a predetermined range, cessation of the monitored person's heartbeat or breathing, stoppage of operation of each of the sensors 21 to 25 or the sensor I / F device 20, operation of the call button 26, etc. Furthermore, in the information processing system, if a personal authentication terminal using face authentication or ID tags is installed at the entrance of the nursing care facility or the door of each monitored person's room, etc., events may include detection of pre-registered persons (monitored person and care staff), detection of unregistered persons, and the monitored person wandering away. The occurrence of an event is determined by the control unit 11 of the server 10 based on information acquired from the sensors 21 to 25 and the call button 26 via the sensor I / F device 20. When the control unit 11 determines that an event has occurred for each monitored person, it registers information about the determined event in the event DB 13c.

[0029] The sensor data DB 13b is provided for each monitoring target and stores the detection results of the sensors 21-25 and operation information for the call button 26 in association with the target ID of the monitoring target. The sensor data DB 13b includes a date and time column, a sleep sensor column, a toilet sensor column, a door sensor column, an environmental sensor column, a human presence sensor column, and a call button column. The date and time column stores the detection date and time of each sensor 21-25 and the operation date and time for the call button 26. The date and time column may store the date and time when the server 10 acquired the detection results and operation information from each sensor 21-25 and the call button 26. The sleep sensor column stores sleep data indicating the sleeping state of the monitoring target detected by the sleep sensor 21. The sleep data includes data indicating the state of getting in and out of bed, body movement data indicating the magnitude of body movement, apnea data indicating the occurrence of apnea, sleep depth data indicating the depth of sleep, heart rate data indicating the heart rate, respiratory rate data indicating the respiratory rate, and body temperature data indicating the body temperature. In addition to storing each piece of sleep data in association with date and time, graph data showing changes in the state of the monitored person over time may also be stored. For example, the sleep sensor array may store graphs showing changes in bed and room conditions over time, graphs showing changes in body movement over time, graphs showing changes in the occurrence of apnea, graphs showing changes in sleep depth over time, graphs showing changes in heart rate over time, graphs showing changes in respiratory rate over time, and graphs showing changes in body temperature over time. The graph data may be a series of data, for example, pairs of the time when the sleep sensor 21 performed a measurement and the measured value, recorded each time a measurement is performed. Note that apnea is a phenomenon in which airflow through the mouth and nose stops for 10 seconds or more during sleep. Sleep depth is detected, for example, in three stages: awake, light sleep, and deep sleep, and the sleep depth data may include the occurrence of mid-sleep awakenings and the total daily sleep time.

[0030] The toilet sensor array stores information (used / not used) indicating whether the toilet sensor 22 has detected the use of the toilet by the person being monitored. The door sensor array stores information (passed / no movement) indicating whether the door sensor 23 has detected the opening of the door or the person being monitored passing through the door. The environmental sensor array stores environmental data indicating the temperature and humidity in the room of the person being monitored detected by the environmental sensor 24. The environmental data may also be stored in the environmental sensor array in the form of graphs showing changes in temperature and humidity over time, in addition to a configuration in which temperature and humidity are stored in association with date and time. The human sensor array stores information (presence / absence) indicating whether the human sensor 25 has detected the presence (presence) of the person being monitored. The call button array stores information (used / not used) indicating whether the person being monitored has operated (used) the call button 26. The control unit 11 of the server 10 updates each column of the sensor data DB 13b based on information acquired from the sensors 21-25 and the call button 26 via the sensor I / F device 20. Note that the toilet sensor column may store only information indicating toilet use (use), the door sensor column may store only information indicating passage through a door (passage), the human sensor column may store only information indicating the presence of a monitored person (presence), and the call button column may store only information indicating operation of the call button 26 (use).

[0031] The event DB 13c is provided for each monitored person and stores information on events that occur to the monitored person in association with the monitored person's ID. The event DB 13c includes a date and time column, an event item column, an event content column, a conversation text column, a conversation voice column, and a progress information column. The date and time column is the date and time when an event occurred, and stores the date and time when the control unit 11 of the server 10 determined that an event occurred based on information acquired from each of the sensors 21-25 and the call button 26. The event item column and the event content column store items related to the event that occurred and their contents. For example, if an event occurs in which a monitored person gets out of bed due to waking up during the night, "getting out of bed" is registered as the event item and "waking up during the night" is registered as the event content. If an event occurs in which the room temperature is outside a predetermined range, "room temperature" is registered as the event item and "30°C (measured room temperature)" is registered as the event content. If an event occurs in which the person stays in the toilet for a long time, "toilet" is registered as the event item and "20 minutes (measured time)" is registered as the event content. Furthermore, if an event such as an increase in body movement during bedtime occurs, "body movement" is registered as the event item, and "increase" is registered as the event content. Furthermore, if an event such as the operation of the call button 26 occurs, "call" is registered as the event item. The conversational voice sequence stores voice data of the conversation between the care staff regarding the event, and the conversational text sequence stores text data obtained by converting the voice data of the conversation into text. The conversational voice data and text data include the date and time when the staff terminal 30 transmitted the voice data and information about the care staff member who spoke. The progress information sequence stores information indicating the progress of the care staff member's response (hereinafter referred to as care) to the event that occurred. For example, if care has been completed, "completed" is stored. The event DB 13c is not limited to the configuration shown in FIG. 3. For example, the event DB 13c may be configured to store, in association with the event item and content, photographs and videos taken by the care staff member when providing care to the monitored person regarding the event, as well as recorded audio.

[0032] When the control unit 11 of the server 10 detects the occurrence of an event for a monitored person based on information acquired from the sensors 21-25 and the call button 26 via the sensor I / F device 20, it updates the date / time column, the event item column, and the event content column in the event DB 13c. The control unit 11 also acquires a conversation exchanged between the staff terminals 30. When the acquired conversational voice is converted into text data, the control unit 11 stores the generated conversational text in a conversational text column and the conversational voice in a conversational voice column. The event DB 13c stores information on all events that have occurred for the monitored person, but only events registered in the monitored person DB 13a that are subject to notification are notified to the care staff. Therefore, the conversational text column and the conversational voice column in the event DB 13c store text and voice related to conversations between the care staff regarding events notified to the care staff. Alternatively, the event DB 13c may be configured such that only information on events notified to the care staff (events registered as subject to notification events) is registered in the event DB 13c.

[0033] The following describes the processing performed by each device in the information processing system of this embodiment. FIGS. 4 and 5 are flowcharts showing an example of a processing procedure in the information processing system. In FIGS. 4 and 5, the left side shows the processing performed by the server 10, and the right side shows the processing performed by the staff terminal 30. The following processing is executed by the control unit 11 in accordance with a program 13P stored in the auxiliary storage unit 13 of the server 10, and by the control unit 31 in accordance with a program 33P and a conversation application 33AP stored in the auxiliary storage unit 33 of the staff terminal 30. In the following processing, when the sensor I / F device 20 acquires a detection signal from each of the sensors 21 to 25 and an operation signal from the call button 26, the sensor I / F device 20 transmits the acquired signals to the server 10 via the network N. Note that the sensor I / F device 20 may be configured to transmit signals from each of the sensors 21 to 25 and the call button 26 to the server 10 in response to a signal request from the server 10.

[0034] The control unit 11 of the server 10 determines whether a signal (sensor data) has been received from the sensors 21 to 25 or the call button 26 connected to the sensor I / F device 20 via any of the sensor I / F devices 20 (S11). If the control unit 11 determines that sensor data has not been received (S11: NO), it waits until it is received. If the control unit 11 (sensor data acquisition unit) determines that sensor data has been received (S11: YES), it identifies which monitored person's state has been detected by the received sensor data (S12). For example, the server 10 is configured to store a table in which identification information (subject ID) of each monitored person is registered in association with identification information (device ID) of each sensor I / F device 20, and the sensor I / F device 20 transmits the sensor data and the identification information of the sensor I / F device 20 to the server 10. In this case, the control unit 11 of the server 10 can identify the monitoring target person corresponding to the received sensor data based on the table and the identification information of the sensor I / F device 20 received from the sensor I / F device 20. The sensor I / F device 20 may also be configured to store the identification information of the corresponding monitoring target person and transmit the sensor data and the identification information of the monitoring target person to the server 10. In this case, the control unit 11 of the server 10 can identify the monitoring target person for the received sensor data based on the identification information of the monitoring target person received from the sensor I / F device 20.

[0035] The control unit 11 stores the date and time at this point in time in the sensor data DB 13b of the identified monitoring target in association with the received sensor data (S13). The control unit 11 (determination unit) identifies an event that has occurred for the monitoring target based on the sensor data stored in the sensor data DB 13b (S14). For example, when a signal indicating that the monitoring target has left bed is received from the sleep sensor 21, the control unit 11 identifies the occurrence of an event that the monitoring target has left bed. Furthermore, when a signal indicating a body movement state is received from the sleep sensor 21, the control unit 11 identifies the occurrence of an event that the monitoring target has made large body movements in bed if the magnitude of the body movement indicated by the received signal is equal to or greater than a predetermined value. Furthermore, when a signal indicating toilet use is received from the toilet sensor 22, the control unit 11 identifies the occurrence of an event that the monitoring target has used the toilet. Furthermore, when the control unit 11 receives a signal indicating toilet use continuously for a predetermined period of time or longer, the control unit 11 identifies the occurrence of an event that the monitoring target has stayed in the toilet for a long time. Furthermore, when a signal indicating a door opening is received from the door sensor 23, the control unit 11 identifies the occurrence of an event that the door opening. Furthermore, when the control unit 11 receives a signal indicating temperature (room temperature) from the environmental sensor 24, if the received temperature is above a predetermined range, it identifies the occurrence of an event of a high temperature state, and if the received temperature is below the predetermined range, it identifies the occurrence of an event of a low temperature state. Furthermore, when the control unit 11 receives a signal indicating humidity from the environmental sensor 24, if the received humidity is above a predetermined range, it identifies the occurrence of an event of a high humidity state (humid state), and if the received humidity is below the predetermined range, it identifies the occurrence of an event of a low humidity state (dry state). Furthermore, when the control unit 11 receives a signal indicating a zero heart rate and a zero respiratory rate from the sleep sensor 21, it identifies the occurrence of an event of an inability to detect a biological reaction (cessation of heart rate and breathing). Furthermore, when the control unit 11 receives an operation signal from the call button 26, it identifies the occurrence of an event of an operation of the call button 26. Furthermore, when the control unit 11 has not received a signal from any of the sensor I / F devices 20 or a signal from any of the sensors 21 to 26 for a predetermined period of time or longer, it identifies the occurrence of an event of a malfunction of any of the sensor I / F devices 20 or any of the sensors 21 to 25.

[0036] Furthermore, if personal authentication terminals using facial recognition or ID tags, etc., are installed at appropriate locations in the nursing care facility, the control unit 11 identifies the occurrence of events such as the detection of a registered person and the detection of an unregistered person based on signals received from the personal authentication terminals. Furthermore, if one personal authentication terminal detects the same monitored person a predetermined number of times or more within a predetermined time, the control unit 11 identifies the occurrence of an event such as the monitored person wandering. The events detected by the control unit 11 are not limited to the events described above, but may be any events occurring in or around the monitored person, and may include various conditions that can be identified based on detection signals from the sensors 21-26. Reference values ​​for determining whether body movement, time spent in the toilet, temperature, and humidity are within normal ranges are pre-stored in the auxiliary storage unit 13 of the server 10, and reference values ​​set for each monitored person may be pre-stored.

[0037] The control unit 11 stores the detection date and time of the sensor data used to identify the occurrence of an event in the event DB 13c of the monitored person in association with information about the determined event (S15). For example, when the control unit 11 identifies the occurrence of an event in which the monitored person gets out of bed due to waking up during sleep, it stores "getting out of bed" as the event item and "waking up during sleep" as the event content. When the control unit 11 identifies the occurrence of an event in which body movement is large, it stores "body movement" as the event item and "large" or "increased" as the event content. When the control unit 11 identifies the occurrence of an event in which the person uses the toilet, it stores "toilet" as the event item. When the control unit 11 identifies the occurrence of an event in which the person stays in the toilet for a long time, it stores the time spent in the toilet as the event content. When the control unit 11 identifies the occurrence of an event in which the person opens a door, it stores "door" as the event item and "open" as the event content. Furthermore, when the control unit 11 identifies the occurrence of an event such as a high temperature or low temperature, it stores "room temperature" as the event item and stores the measured room temperature as the event content. When the control unit 11 identifies the occurrence of an event such as a humid or dry temperature, it stores "humidity" as the event item and stores the measured humidity as the event content. When the control unit 11 identifies the occurrence of an event such as cessation of heartbeat and breathing, it stores "heart rate" as the event item and stores the measured heart rate and respiratory rate (measured number zero) as the event content. When the control unit 11 identifies the occurrence of an event such as operation of the call button 26, it stores "call button" as the event item and stores "operation" as the event content. When the control unit 11 identifies the occurrence of an event such as malfunction of the sensor I / F device 20 or the sensors 21 to 25, it stores "device malfunction" as the event item and stores device information of the device in which the occurrence of unnecessary operation was detected as the event content.

[0038] Furthermore, when the control unit 11 identifies the occurrence of an event such as the detection of a registered person or the detection of an unregistered person, it stores "person detection" as the event item and stores information indicating the detected monitored person or unregistered person as the event content. Furthermore, when the control unit 11 identifies the occurrence of an event such as the monitored person wandering, it stores "wandering" as the event item and stores information on the monitored person detected as the event content. Note that if the personal authentication terminal is installed in a common area for the monitored person, an event DB for the nursing care facility may be prepared in the server 10, and event information indicating the detection of a registered person, the detection of an unregistered person, and the monitored person wandering, as well as identification information or installation location information of the personal authentication terminal, may be stored in the event DB for the nursing care facility.

[0039] The control unit 11 determines whether the event identified in step S14 is an event to be notified to the care staff (S16). Here, the control unit 11 determines whether the event identified in step S14 is included in the events to be notified set for the monitoring target person identified in step S12. If it is determined that the event is not an event to be notified (S16: NO), the control unit 11 ends the processing. Note that the control unit 11 executes the processing from step S12 onwards every time it receives sensor data from the sensor I / F device 20. As a result, the server 10 accumulates the sensor data received from the sensor I / F device 20 in the sensor data DB 13b of each monitoring target person, and accumulates information on events that have occurred to each monitoring target person in the event DB 13c.

[0040] If it is determined that the event is one requiring notification (S16: YES), the control unit 11 generates alert information to be notified to the care staff (S17). The alert information includes the content of the event requiring notification and information about the monitored person for whom the event occurred. For example, in the case of an event in which a monitored person leaves bed, the control unit 11 generates alert information including the room number and name of the monitored person and a message indicating that the monitored person has left bed. Specifically, alert information such as "Room 503, Taro Yamada has left bed" is generated. In the case of an event in which a person is making large body movements, the control unit 11 generates alert information including the room number and name of the monitored person and a message indicating that the monitored person is making large body movements. In the case of an event in which a person stays in the toilet for a long time, the control unit 11 generates alert information including the room number and name of the monitored person and a message indicating the length of time the person stayed in the toilet. In the case of an event in which a door is opened, the control unit 11 generates alert information including the room number and name of the monitored person and a message indicating that the door is opened. Furthermore, in the event of a high temperature or low temperature, the control unit 11 generates alert information including the room number and name of the person being monitored and a message indicating the room temperature. Furthermore, in the event of a humid or dry condition, the control unit 11 generates alert information including the room number and name of the person being monitored and a message indicating the humidity. Furthermore, in the event of a cardiac arrest or respiratory arrest, the control unit 11 generates alert information including the room number and name of the person being monitored and a message indicating that the cardiac arrest or respiratory arrest may have occurred. Furthermore, in the event of an operation of the call button 26, the control unit 11 generates alert information including the room number and name of the person being monitored and a message indicating that the call button 26 has been operated. Furthermore, in the event of a malfunction of the sensor I / F device 20 or the sensors 21 to 25, the control unit 11 generates alert information including the room number and name of the person being monitored and a message indicating that the equipment is malfunctioning.

[0041] Next, the control unit 11 (output unit) transmits the generated alert information to the staff terminal 30 via the network N (S18). For example, the control unit 11 transmits the alert information to all staff terminals 30 with which it can communicate by broadcast communication. This allows the server 10 to output the occurrence of an event in the monitored person to multiple staff terminals 30. The control unit 11 may store the alert information in the event DB 13c in association with event information. The control unit 31 of the staff terminal 30 receives the alert information transmitted by the server 10 via the communication unit 34 and outputs the received alert information as audio via the earphones 36 (S19). The control unit 11 of the server 10 may generate alert information as text data, convert the generated alert information into audio data, and transmit the generated audio data alert information to the staff terminal 30. In this case, the control unit 31 of the staff terminal 30 can output the received audio data alert information as audio via the earphones 36. The control unit 11 of the server 10 may also generate alert information as text data and transmit the generated text data alert information to the staff terminal 30. In this case, the control unit 31 of the staff terminal 30 converts the received text data alert information into audio data alert information, and can output the obtained audio data alert information aloud from the earphones 36. For example, the control unit 31 outputs the alert information aloud, "Room 503, Yamada Taro has left his bed." Through the above-described processing, when the server 10 detects the occurrence of an event that has been set as a notification target for each monitored person, it can notify the care staff of the occurrence of the event via the staff terminal 30.

[0042] When a care staff member is notified of the occurrence of an event via the staff terminal 30, the care staff member uses the staff terminal 30 to converse with other care staff members and provide care (necessary treatment) for the event that has occurred. The control unit 31 of the staff terminal 30 determines whether or not it has acquired voice data of a speech (conversational voice) uttered by the care staff member who is the user of the staff terminal 30 via the microphone 35 (S20). If it is determined that the speech voice of the care staff member has been acquired (S20: YES), the control unit 31 transmits the acquired voice data to the other staff terminals 30 and the server 10 via the network N (S21).

[0043] If the control unit 31 determines that the voice of the care staff member has not been acquired from the staff terminal 30 (S20: NO), the control unit 31 proceeds to the process of step S22. That is, the control unit 31 of the staff terminal 30 that has not acquired the voice of the care staff member determines whether or not it has received voice data transmitted from another staff terminal 30 that has acquired the voice of the care staff member (S22). If the control unit 31 determines that it has received voice data transmitted from another staff terminal 30 (S22: YES), it outputs the conversation voice of the other care staff member through the earphone 36 based on the received voice data (S23). If the control unit 31 determines that it has not received voice data from another staff terminal 30 (S22: NO), it skips the process of step S23 and returns to the process of step S20, and repeats the process of transmitting the acquired voice data to the other staff terminal 30 and the server 10 every time it acquires the voice of the care staff member through the microphone 35. Furthermore, the control unit 31 repeats the process of outputting conversational voice based on the received voice data from the earphones 36 each time it receives voice data transmitted from another staff terminal 30. This allows the care staff to converse with each other via the staff terminal 30. For example, if care staff member A says, "Yamada-san in 503, you've gotten out of bed. Is anyone nearby?", the staff terminal 30 of care staff member A acquires the voice data transmitted by care staff member A and transmits it to the other staff terminal 30. The other staff terminal 30 outputs the received voice data. If care staff member B, hearing this voice, utters, "It's B. I'm nearby, so I'll help you now," the staff terminal 30 of care staff member B acquires the voice data transmitted by care staff member B and transmits it to the other staff terminal 30. The other staff terminal 30 outputs the received voice data. If care staff member A, hearing this voice, utters, "Mr. B, please," the staff terminal 30 of care staff member A acquires the voice data transmitted by care staff member A and transmits it to the other staff terminal 30. The other staff terminals 30 output the received voice data as voice, and each care staff member who hears this voice can understand that care staff member B will provide care in response to the notified event (alert).

[0044] If care staff member B needs help or advice from other care staff members while providing care in response to the alert, he / she will emit a conversational voice requesting help or advice. In this case, the staff terminal 30 of care staff member B can request help or advice from the other care staff members by acquiring the voice data emitted by care staff member B and transmitting it to the other staff terminals 30. Furthermore, when care staff member B completes care for the notified event, he / she will emit a conversational voice reporting the completion of care. In this case, the staff terminal 30 of care staff member B can report the completion of care to the other care staff members by acquiring the voice data emitted by care staff member B and transmitting it to the other staff terminals 30. In this way, each care staff member can provide care while conversing with each other via the staff terminal 30.

[0045] The control unit 11 (voice data acquisition unit) of the server 10 receives the voice data transmitted from each staff terminal 30 (S24). The control unit 31 of the staff terminal 30 transmits the voice data and the identification information of the staff terminal 30 itself or the identification information of the care staff member using the staff terminal 30 itself, and when the control unit 11 of the server 10 receives the voice data, it can identify the staff terminal 30 that transmitted the voice data or the care staff member using the staff terminal 30. The control unit 11 (conversion unit) converts the received voice data into text data and generates text data of the conversation voices of the care staff members (S25).

[0046] The control unit 11 then stores the text data generated in step S25 and the voice data received in step S24 in the event DB 13c (storage unit) for the current monitoring target (the monitoring target for whom an event is occurring) (S26). Specifically, the control unit 11 associates the identification information of the care staff member who generated the received voice data, the date and time of reception of the voice data, and the text data of the generated conversational voice and stores them in a conversation text string, and associates the identification information of the care staff member, the date and time of reception of the voice data, and the voice data received in step S24 in a conversational voice string. The control unit 11 stores the conversation text and the conversational voice in the event DB 13c in association with the event stored in the event DB 13c in step S15. Through the above-described process, the control unit 11 can acquire voice data of the conversation between the care staff members from each staff terminal 30 and can store an event that occurred to the monitoring target and the text data and voice data of the conversation between the care staff members regarding the event in association with each other.

[0047] The control unit 11 determines whether the care provided by the care staff has ended for the event notified to the care staff by sending alert information to the staff terminal 30 in step S18 (S27). For example, the staff terminal 30 may be provided with an input unit (input button) for inputting the end of care, and the control unit 11 may determine the end of care when the input button is operated. The control unit 11 may also determine whether the care has ended based on whether a predetermined keyword is included in the voice data uttered by the care staff while providing care. For example, keywords such as "end" or "end (completion)" may be registered in advance in the auxiliary storage unit 13, and the control unit 11 may determine that the care has ended if the keyword is included in the voice data uttered by the care staff. Note that in addition to determining whether the keyword is included in the voice data uttered by the care staff, the control unit 11 may also determine whether the keyword is included in text data converted from the voice data.

[0048] The end of nursing care by the care staff may also be determined using a learning model trained by machine learning. The learning model is expected to be used as a program module that functions as part of artificial intelligence software. For example, a learning model trained using algorithms such as CNN (Convolution Neural Network), RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), Transformer, BERT (Bidirectional Encoder Representations from Transformers), and GPT (Generative Pre-Training) can be used to output information indicating whether conversation data (voice data or text data) by the care staff contains content related to nursing care for the monitored individual when the conversation data is input. In this case, the control unit 11 inputs the conversation data of the care staff into the learning model, determines whether the conversation data contains content related to nursing care by the care staff based on the output information from the learning model, and determines whether nursing care by the care staff has ended based on the determination result.

[0049] FIG. 6 is an explanatory diagram showing an example of the configuration of a learning model. The learning model shown in FIG. 6 is trained to receive input data (audio data or text data) of a conversation between a care staff member using the staff terminal 30, perform a calculation based on the input data to determine whether the conversation is related to care provided by the care staff member, and output the calculation result. The learning model shown in FIG. 6 has two output nodes. Output node 0 outputs the probability that the input conversation data should be determined to include content related to care provided by the care staff member, and output node 1 outputs the probability that the input conversation data should be determined to include content other than that, such as casual conversation. The output value of each output node is, for example, a value between 0 and 1.0, and the sum of the determination probabilities output from each output node is 1.0.

[0050] The learning model shown in FIG. 6 is generated by machine learning an untrained learning model using training data containing training conversation data (voice data or text data) and information (correct labels) indicating whether the conversation data includes content related to nursing care or other content (chat). When training conversation data is input, the learning model learns so that the output value from the output node corresponding to the content indicated by the correct label approaches 1.0 and the output values ​​from other output nodes approach 0.0. In the learning process, the learning model performs calculations based on the input conversation data to calculate output values ​​from each output node. The learning model then compares the calculated output values ​​of each output node with the value corresponding to the correct label (0 or 1) and optimizes parameters used in the calculation process so that each output value approximates the value corresponding to the respective correct label. The parameters are, for example, weights between neurons in the learning model. The parameter optimization method is not particularly limited, but examples include backpropagation, steepest descent, and the like. This results in a learning model that is trained to determine whether, when conversation data is input, the input conversation data contains content related to nursing care or is casual conversation that does not contain content related to nursing care.

[0051] The learning model is trained by another learning device. The trained learning model generated by the other learning device is downloaded from the learning device to the server 10 via the network N or the portable storage medium 10a, for example, and stored in the auxiliary storage unit 13. Note that the learning model is not limited to the configuration shown in FIG. 6, and the number of discrimination targets determined by the learning model is not limited to two. For example, the learning model may be configured to determine whether the input conversation data includes content related to nursing care, casual conversation, or content indicating the end of nursing care. In this case, the control unit 11 can determine whether nursing care by the nursing staff has ended based on the output information from the learning model. Furthermore, the learning model may be configured to determine whether the input conversation data includes content related to nursing care, casual conversation, content indicating the start of nursing care, content indicating that nursing care is being provided, or content indicating the end of nursing care. In this case, the progress status of the notified event, such as whether nursing care by the nursing staff has started, is being provided, or has ended, can be determined.

[0052] Returning to the explanation of the process shown in Fig. 5, if it is determined that the nursing care by the nursing staff has not ended (S27: NO), the control unit 11 returns to the process of step S24 and executes the processes of steps S24 to S27 on the voice data transmitted from the staff terminal 30. This allows the voice data of the conversation transmitted from the staff terminal 30 and the text data of the conversation generated from the voice data to be stored in the event DB 13c. Therefore, the voice data and text data of the conversation conducted by the nursing staff using the staff terminal 30 are stored in the event DB 13c.

[0053] When it is determined that the care provided by the care staff has ended (S27: YES), the control unit 11 stores "Done" in the progress information field of the event DB 13c for the monitored person for whom care has been completed, indicating that the care has been completed (S28), and ends the series of processes. Through the above-described processes, the server 10 can determine the status of each monitored person based on signals from the sensors 21-25 and the call button 26 associated with each monitored person, and can detect the occurrence of an event that requires notification to the care staff. Furthermore, when the server 10 detects the occurrence of an event for each monitored person, it can notify each of the monitored people via the staff terminal 30. The care staff can use the staff terminal 30 to converse with other care staff, and the server 10 can store text data and audio data of the conversations between the monitored people in association with the event.

[0054] The text data of conversations between care staff members stored in the event DB 13c by the above-described process can be viewed, for example, using the staff terminal 30 or another terminal. FIG. 7 is an explanatory diagram showing an example of display of the text data of conversations. When viewing the text data of conversations stored in the event DB 13c, the viewer specifies an arbitrary monitored person and an arbitrary date and time or an event that has occurred, and requests the text data of the conversation corresponding to the specified event from the server 10. The control unit 11 of the server 10 reads out information about the event corresponding to the requested content and the text data of the conversation (conversation text) from the event DB 13c, and transmits them to the requesting terminal.

[0055] When the requesting terminal receives the event information and the text data of the conversation from the server 10, it generates a screen as shown in Fig. 7A and displays it on the display unit. In the screen shown in Fig. 7A, the alert information generated by the control unit 11 of the server 10 in step S17 of Fig. 4 and the date and time of the alert (event) occurrence are displayed in a bold-line display frame. In addition, the information of the care staff who spoke, the date and time of the speech, and the text data of the conversation between the care staff generated by the control unit 11 of the server 10 in step S25 of Fig. 5 are displayed in a thin-line display frame in chronological order, each associated with the other. From the screen as shown in Fig. 7A, the viewer can understand the content of the event that occurred, the content of the conversation between the care staff regarding the event, and the content of the care provided to the monitored person.

[0056] In this embodiment, multiple types of sensors 21 to 25 are provided to determine the state of the person being monitored, and the state of the person being monitored (occurrence of an event) is monitored based on detection signals from each of the sensors 21 to 25. Furthermore, for each person being monitored, it is possible to set events that should be notified to the care staff among the events that have occurred. Therefore, different events can be set as events that should be notified to the care staff depending on the health condition, etc. of the person being monitored.

[0057] In this embodiment, when generating text data from audio data of conversations between care staff members received from the staff terminal 30, the control unit 11 of the server 10 may be configured to extract pre-registered keywords from the generated text data and perform highlighting to highlight the extracted keywords. For example, after processing step S25 in FIG. 5, the control unit 11 identifies predetermined keywords from the generated text data and performs highlighting on each keyword. In this configuration, when viewing the text data of conversations stored in the event DB 13c of the server 10, predetermined keywords are highlighted as shown in FIG. 7B. This allows the viewer to grasp important keywords from the content of conversations between multiple care staff members without missing them. In the example shown in FIG. 7B, the room number and name of the monitored person, the content of the event, and the content of the care are displayed in a frame.

[0058] In the process shown in FIGS. 4 and 5 of this embodiment, the server 10 may be configured to, when acquiring a conversational voice of a care staff member from the staff terminal 30, perform a process of identifying which event occurred in which monitored person the conversational voice is uttered in relation to. The control unit 11 of the server 10 performs the processes of steps S12 to S18 each time sensor data is received from each sensor I / F device 20. Therefore, the server 10 may transmit alert information related to different events consecutively to the staff terminal 30. That is, a situation may arise in which, after alert information related to one event is transmitted to the staff terminal 30, alert information related to another event is transmitted to the staff terminal 30 before the care staff member finishes providing care for the event. In this case, the care staff member will be conversing about multiple events, and it becomes necessary to determine which event each conversation of the care staff member relates to. Therefore, for example, after processing step S25 in Figure 5, the control unit 11 of the server 10 identifies an event that corresponds to the voice data (voice of the care staff conversation) received in step S24 from among the events being notified to the care staff (events for which care by the care staff has not yet been completed).

[0059] For example, the control unit 11 determines that a conversation voice received within a predetermined time after the alert information is transmitted is a conversation corresponding to the event of the transmitted alert information. Specifically, the control unit 11 determines that a conversation voice received from the staff terminal 30 within a predetermined time (e.g., within 30 seconds) after the alert information is transmitted in step S18, and a conversation voice of another care staff member received within a predetermined time (e.g., within 10 seconds) after the reception of the conversation voice, are conversations corresponding to the event for which the alert information is transmitted in step S18. This allows a conversation that took place within a predetermined time after the notification of the occurrence of an event to be identified as a conversation related to the event. The control unit 11 may also determine, from the received conversation content, which event occurred to which monitored person the conversation is related to. For example, if the conversation content includes information about the monitored person or information about an event, the control unit 11 can identify the event corresponding to the received conversation voice based on the information included in the conversation content. In this case, even if a predetermined time or more has passed since the transmission of the alert information, it is possible to identify which event the conversation voice of the care staff member is related to. If there is only one event for which the care staff is being notified, the control unit 11 may determine that the voice data received in step S24 is a conversation voice relating to the event for which the alert information was transmitted in step S18.

[0060] Then, in step S26, the control unit 11 stores the text data of the conversation voice of the care staff in the conversation text string corresponding to the identified event in the event DB 13c of the monitored person for whom the event has occurred, and stores the audio data of the conversation voice of the care staff in the conversation audio string corresponding to the identified event. This allows the conversation voice and conversation text of the care staff to be stored in association with the corresponding event.

[0061] In the above-described embodiment, the server 10 is configured to convert the acquired voice data into text data each time the server 10 acquires voice data of a conversation between care staff using the staff terminal 30, and store the converted voice data in the event DB 13c. However, the process of converting voice data into text data does not need to be performed at the timing when the voice data is acquired, and may be performed, for example, at predetermined time intervals. For example, when the server 10 acquires voice data of a conversation between care staff, the server 10 may store the acquired voice data in the event DB 13c, and at predetermined times, convert each voice data into text data and store the text data in the event DB 13c.

[0062] (Embodiment 2) This section describes an information processing system in which, when the server 10 acquires conversational voices of care staff from the staff terminal 30, it determines whether the conversational voices are conversations about responses (care) to an event that has occurred to a monitored person, and stores information about the conversations about responses. The information processing system of this embodiment can be realized using devices similar to the information processing system of embodiment 1 shown in Fig. 1, and therefore a description of the configuration of each device will be omitted.

[0063] Fig. 8 is a flowchart showing an example of a processing procedure in the information processing system of embodiment 2. The processing shown in Fig. 8 is obtained by adding steps S31 and S32 between steps S25 and S26 in the processing shown in Fig. 4 and Fig. 5. Explanations of the same steps as in Fig. 4 and Fig. 5 will be omitted. Also, in Fig. 8, illustration of each step in Fig. 4 is omitted.

[0064] The server 10 and staff terminal 30 of this embodiment perform the same processes as steps S11 to S25 in Figures 4 and 5. As a result, the server 10 acquires signals from the sensors 21 to 25 and the call button 26 via each sensor I / F device 20, and detects the occurrence of an event for each monitored person based on the acquired signals. Furthermore, when the server 10 detects the occurrence of an event that should be notified to the care staff, it notifies the care staff of the occurrence of the event via the staff terminal 30. Furthermore, when the server 10 acquires the spoken voice of the care staff from the staff terminal 30, it converts the acquired voice data into text data.

[0065] After processing step S25, the control unit 11 of the server 10 of this embodiment identifies the content of the conversational voice received from the staff terminal 30 (S31). For example, the control unit 11 identifies whether the content of the conversational voice is related to nursing care provided by the nursing staff to the monitored person or is just chatting. For example, if the conversational voice includes information about the monitored person or information about an event, the control unit 11 determines that the conversational voice is related to nursing care provided by the monitored person. Alternatively, if the conversational voice does not include information about the monitored person or information about an event, the control unit 11 may determine that the conversational voice is not related to nursing care provided by the monitored person but is, for example, just chatting. Alternatively, the control unit 11 may use a learning model to determine whether the content of the conversational voice is related to nursing care provided by the monitored person or just chatting. For example, the learning model shown in FIG. 6 can be used. In this case, the control unit 11 inputs the voice data received from the staff terminal 30 (or text data generated from the voice data) into the learning model and, based on output information from the learning model, determines whether the content of the received conversational voice is related to nursing care provided by the nursing staff or just chatting.

[0066] The control unit 11 determines whether the conversation received from the staff terminal 30 is related to nursing care for the monitored person based on the identified conversation content (S32). If it is determined that the conversation is related to nursing care for the monitored person (S32: YES), the control unit 11 stores the text data generated in step S25 and the voice data received in step S24 in the event DB 13c of the monitored person for whom the event is occurring (S26). On the other hand, if it is determined that the conversation is not related to nursing care for the monitored person (S32: NO), for example, if it is just casual conversation, the control unit 11 skips the processing of step S26.

[0067] When the content of the conversation between the care staff is determined to be casual conversation through the above-described process, the text data and voice data of the conversation are not stored in the event DB 13c, thereby preventing the accumulation of conversation information of the care staff's casual conversation in the event DB 13c. Thus, the server 10 can extract conversation information related to nursing care by the care staff from the conversation information (conversation voice and conversation text) by the care staff and store the extracted conversation information in association with the event that occurred. Note that in the above-described process, both the conversation information related to nursing care by the care staff and the conversation information of the care staff's casual conversation may be stored in the event DB 13c, and each piece of conversation information may be stored in association with information indicating whether the content is related to nursing care or casual conversation.

[0068] In this embodiment, the same effects as those of the first embodiment can be obtained. Furthermore, in this embodiment, among the conversation information by the care staff received by the server 10 from the staff terminal 30, conversation information that needs to be accumulated, such as conversations about care provided to the monitored person, can be accumulated. Therefore, when checking the care provided by the care staff for each event, it becomes possible to efficiently view the conversation information about the care provided by the care staff. In this embodiment, the modified examples described as appropriate in the first embodiment can also be applied.

[0069] (Embodiment 3) This section describes an information processing system in which a server 10 creates a report on an event that occurred to a monitored person based on the conversation content (conversation text) between care staff members that is accumulated in response to the event. The information processing system of this embodiment can be realized using devices similar to those of the information processing system of embodiment 1 shown in Fig. 1, and therefore a description of the configuration of each device will be omitted.

[0070] In the information processing system of this embodiment, the server 10 and the staff terminal 30 execute the same processes as those shown in Figures 4 and 5. As a result, the server 10 can detect the occurrence of an event for each monitored person based on signals from the sensors 21 to 25 and the call button 26 acquired via each sensor I / F device 20, and transmit alert information to each staff terminal 30 as necessary. Furthermore, when the server 10 notifies the care staff of alert information (occurrence of an event), it can convert the spoken voice of the care staff acquired from the staff terminal 30 into text data and store it in association with the event.

[0071] The server 10 of this embodiment has the function of automatically creating various types of report information (reports), such as information to be accumulated as a care record (care record information), information to be reported to medical professionals such as doctors (doctor report information), and information to be reported to family members of the monitored person (family report information), based on information regarding events that occurred to the monitored person that was accumulated in the event DB 13c by the processing shown in Figures 4 and 5.

[0072] FIG. 9 is a flowchart showing an example of a procedure for creating report information, and FIG. 10 is an explanatory diagram showing an example of report information. The following process is executed by the control unit 11 in accordance with a program 13P stored in the auxiliary storage unit 13 of the server 10. The control unit 11 of the server 10 is configured to periodically execute the following process and starts executing the process from step S41 onwards when a predetermined execution timing arrives. The predetermined execution timing may be, for example, when a predetermined time arrives if report information is created once a day, or when a predetermined time arrives on a predetermined day if report information is created once a week. The frequency of creating report information is not limited to once a day or once a week, but may be, for example, multiple times a day (once at a predetermined time) or once every predetermined number of days. It may also be determined that the execution timing has arrived when a creation request is received from a care staff member or the like. Different creation frequencies may be set for the care record information, the doctor report information, and the family report information, and the creation frequency of each piece of report information may be set differently for each monitoring target. In the case where the frequency of creating each piece of report information is set for each person to be monitored, the frequency of creating each piece of report information may be registered in the monitored person DB 13a in association with the person to be monitored.

[0073] When a predetermined execution timing arrives, the control unit 11 of the server 10 determines whether there are any subjects (monitored persons) for whom report information should be created at this time (S41). For example, when the timing for creating each piece of report information arrives, the control unit 11 creates report information corresponding to each of the monitored persons, regarding all monitored persons as persons for whom report information should be created. Furthermore, when the timing for creating report information set for an arbitrary monitored person arrives, the control unit 11 creates report information corresponding to that monitored person, regarding that monitored person as a person for whom report information should be created.

[0074] When it is determined that there is a person for whom report information is to be created (S41: YES), the control unit 11 reads information about events stored in the event DB 13c corresponding to one monitoring target person (S42). The control unit 11 reads information about events that occurred within a predetermined period corresponding to the timing of report information creation, thereby reading information about events for which report information has not yet been created (unreported events). For example, the event DB 13c may be configured to store status information indicating whether report information has been created (reported) in association with information about each event. In this case, the control unit 11 simply reads information about events for which status information has not yet been reported. Note that the control unit 11 reads the date and time, event item, event content, conversation text, and progress information from the event DB 13c as information about each event.

[0075] The control unit 11 identifies the content of one event from the read event information (S43). For example, the control unit 11 identifies the date and time of occurrence and content of the event based on the read date and time, event item, and event content. For example, the control unit 11 identifies "2021 / 9 / 1 1:00 Getting out of bed (waking up mid-day)" as the content of the event. Next, the control unit 11 identifies the content of the care provided by the care staff for the identified event from the read conversation text (S44). For example, the control unit 11 identifies "2021 / 9 / 1 1:05 Care staff A: Toilet assistance" and "2021 / 9 / 1 1:15 Completed response" as the care provided by the care staff. Note that information about care that the care staff may provide may be registered in advance in the auxiliary storage unit 13 of the server 10, and the control unit 11 may identify the care provided by the care staff by extracting the registered information from the conversation text.

[0076] The control unit 11 may also use a learning model to summarize the conversation text and identify the content of the care provided by the care staff. For example, a learning model can be used that has been trained using an algorithm such as CNN, RNN, LSTM, Transformer, BERT, or GPT to summarize the content of the conversation when text data of the conversation is input, and output information about the care provided by the care staff identified from the content of the conversation. In this case, the control unit 11 inputs the conversation text into the learning model, and can identify the care staff who provided care for the event and the care provided by the care staff based on the output information from the learning model.

[0077] The learning model here is a learning model trained to input text data of conversations between care staff and output text data as a summary result obtained by summarizing the input text data. For example, a learning model optimized as parameters for performing general-purpose summarization processing is generated by fine-tuning (transfer learning) a learning model using training data including training text data and words or sentences (correct labels) to be extracted as a summary result of the text data. In fine-tuning, the learning model receives training text data as input, performs calculations based on the input text data, and calculates an output value between 0 and 1 for each word or sentence in the input text data. The learning model then compares the calculated output value for each word or sentence with a value corresponding to the correct label (1 if correct, 0 if incorrect), and optimizes the parameters used in the calculation process so that the two values ​​approximate each other. This results in a learning model that, when text data of conversations between care staff is input, outputs text data as a summary result obtained by summarizing the input text data. The learning model here is trained using another learning device. A trained learning model generated by training on another learning device is downloaded from the learning device to the server 10 via the network N or a portable storage medium 10a, for example, and stored in the auxiliary storage unit 13.

[0078] The control unit 11 determines whether the contents of all events have been identified based on the information read from the event DB 13c in step S42 (S45). If the control unit 11 determines that the contents of all events have not been identified (S45: NO), the control unit 11 returns to the process of step S43 and executes the processes of steps S43 to S44 for the unidentified events. As a result, the control unit 11 identifies the contents of other events from the event information read in step S42 (S43), and identifies the contents of care provided by the care staff for the identified events (S44).

[0079] When the control unit 11 determines that the contents of all events have been identified based on the information read in step S42 (S45: YES), it creates care record information based on the contents of the events identified in step S43 and the contents of the care provided by the care staff identified in step S44 (S46). As shown in FIG. 10, the care record information is created in association with the monitoring target for whom report information is to be created, and includes information for each identified event, such as the date and time and contents of the event, the date and contents of the care provided by the care staff, and the date and time the care provided by the care staff was completed. For example, a format for care record information is stored in advance in the auxiliary storage unit 13, and the control unit 11 generates care record information by inputting the contents of the identified events and the contents of the care provided by the care staff into each field of the format for care record information.

[0080] The control unit 11 also generates report information for the doctor based on the details of the events identified in step S43 and the details of the care provided by the care staff identified in step S44 (S47). The doctor report information is also generated in association with the monitoring target for which the report information is generated. As shown in FIG. 10, the doctor report information includes information such as the date and time of occurrence and details of each identified event, details to be reported to the doctor based on the event, and details to be confirmed with the doctor regarding the report details. The example of the doctor report information shown in FIG. 10 includes a report regarding the sleep state, such as getting up from bed multiple times to use the toilet during the night, and a confirmation requesting the doctor to confirm whether the report details are related to the taking of sleep medication and diuretic medication. The report details and confirmation details for the doctor may be stored in advance in a database provided in the auxiliary storage unit 13 of the server 10, for example, in association with the details and frequency of occurrence of each event. In this case, the control unit 11 can obtain the report details and confirmation details for the doctor from the database according to the details of the event identified in step S43. The contents of the report or confirmation to the doctor may be input by the care staff from the staff terminal 30 or another terminal. In this case, too, a format for the doctor report information is stored in advance in the auxiliary storage unit 13, and the control unit 11 generates the doctor report information by inputting the contents of the identified event, etc., into each field of the format for the doctor report information.

[0081] The control unit 11 also generates report information for the family of the monitored person based on the details of the events identified in step S43 and the details of the care provided by the care staff identified in step S44 (S48). The family report information is also generated in association with the monitored person for whom the report information is generated. As shown in FIG. 10, the family report information includes information about the monitored person's daily life, including the occurrence status of each identified event, the details of the care provided by the care staff in response to the event, and information indicating that the details of the event have been shared with a doctor and that appropriate care is being provided in cooperation with the doctor. The example of family report information shown in FIG. 10 includes information about the monitored person's daily life, such as getting up to go to the toilet multiple times during the night, the details of the care provided by the care staff in response to the events, and information indicating that the doctor has confirmed the relationship between the events and the medications the monitored person is taking. The report information for the family may also be stored in advance in a database provided in the auxiliary storage unit 13 of the server 10, for example, in association with the details and frequency of each event. In this case, the control unit 11 can retrieve the report information for the family from the database according to the details of the event identified in step S43. The content of the report to the family may be input by the care staff from the staff terminal 30 or another terminal. In this case, too, for example, a format for family report information is stored in advance in the auxiliary storage unit 13, and the control unit 11 generates family report information by inputting the content of the identified event and the content of the care provided by the care staff into each field of the format for family report information.

[0082] When generating each piece of report information in steps S46 to S48, the control unit 11 may identify pre-registered keywords from each piece of report information and perform highlighting processing to highlight the identified keywords. The control unit 11 stores the care record information created in step S46, the doctor report information created in step S47, and the family report information created in step S48, for example, in the monitoring target DB 13a in association with the target ID (S49). The control unit 11 may also store the report information created in steps S46 to S48 in the event DB 13c for each monitoring target in association with the date, time, item, and content of each event. The control unit 11 may also transmit the created care record information to a server or terminal that shares care records among care staff, transmit the created doctor report information to a terminal used by the corresponding doctor, or transmit the created family report information to a pre-registered family terminal. The control unit 11 also associates each piece of report information with the created event and stores information indicating that the event has been reported in the event DB 13c. Specifically, the control unit 11 stores information indicating that the event has been reported in the event DB 13c in association with the event whose information has been read out from the event DB 13c in step S42.

[0083] The control unit 11 returns to the process of step S41 and determines whether there is a monitoring subject for whom report information should be created but for whom report information has not yet been created (S41). If it is determined that there is no monitoring subject for whom report information should be created (S41: NO), the control unit 11 terminates the series of processes. Through the above-described process, various types of report information can be automatically created based on the conversation text of the care staff stored by the server 10 in association with events that occurred to the monitoring subject. Furthermore, report information with content appropriate to the report information submission destination, such as other care staff, doctors, or the monitoring subject's family, can be created. Note that if different creation timings are set for the care record information, doctor report information, and family report information, the control unit 11 only needs to generate report information when the creation timing has arrived, and may execute one or more of steps S46 to S48 at an appropriate creation timing. Furthermore, the creation timing of each piece of report information may be set for each monitoring subject. For example, the creation timing of report information for doctors may be changed depending on the monitoring subject's condition, etc., and the creation timing of report information for the monitoring subject's family may be changed depending on the frequency of requests from the family. Furthermore, the report information is not limited to being created periodically, but may be created, for example, when some kind of trouble occurs in the care of the monitored person, in which case the report information makes it possible to clarify how the trouble occurred.

[0084] In this embodiment, in addition to the configuration in which the conversation text is summarized using a learning model to identify the content of the care provided by the care staff, a configuration in which care record information, doctor's report information, and family report information are created from the conversation text using a learning model may also be used. In this case, for example, the care record information may be created using a learning model for care records that has been trained to input text data of a conversation spoken by the care staff and output care record information (text data) generated by summarizing the input text data. Furthermore, the doctor's report information may be created using a learning model for doctor reports that has been trained to input text data of a conversation spoken by the care staff and output doctor's report information (text data) generated by summarizing the input text data. Furthermore, the family report information may be created using a learning model for family reports that has been trained to input text data of a conversation spoken by the care staff and output family report information (text data) generated by summarizing the input text data. Using such learning models, care record information, doctor's report information, and family report information can be created from the conversation data of the care staff, thereby reducing the workload of the care staff.

[0085] In this embodiment, the same effects as those of the above-described embodiments can be obtained. Furthermore, in this embodiment, care record information, report information to a doctor, and report information to a family member of a person being monitored can be automatically created based on text data of conversations between care staff members that the server 10 stores in association with each event. This reduces the workload of care staff members when creating this report information. The configuration of this embodiment can be applied to the information processing systems of the above-described embodiments 1 and 2, and similar effects can be obtained even when applied to the information processing systems of the above-described embodiments. Furthermore, the modified examples described in the above-described embodiments can also be applied to this embodiment.

[0086] (Embodiment 4) An information processing system including a nurse call system and storing information on conversations between a person to be monitored and care staff using the nurse call system will be described below. Fig. 11 is an explanatory diagram showing an example of the configuration of an information processing system according to a fourth embodiment.

[0087] The information processing system of this embodiment includes a server 10, a staff terminal 30, a base unit 40 and a handset 50 in a nurse call system, and the like, and these devices are communicatively connected via a network N. Note that, for example, the handset 50 may be connected to the base unit 40 via a dedicated communication line or wireless communication instead of being connected to the network N. Also, although not shown in FIG. 11 , the information processing system of this embodiment may include a sensor I / F device 20 and various sensors 21 to 26 (see FIG. 1 ) installed for each monitored person in addition to the configuration shown in FIG. 11 . Also, the handset 50 of the nurse call system may be installed in the monitored person's room instead of the call button 26. In this embodiment, the server 10 and the staff terminal 30 are the same devices as the server 10 and the staff terminal 30 in the information processing system of embodiment 1, and therefore description thereof will be omitted.

[0088] The nurse call system includes a handset 50 used as a calling device and a base unit 40 used as a response device that responds to calls from the handset 50. The handset 50 is installed, for example, in each room of the person being monitored or in a shared space, and the base unit 40 is installed, for example, in a room (station) where the care staff is stationed or in an office of a care facility. Note that the staff terminal 30 carried by the care staff may be configured to be usable as a response device for the nurse call system.

[0089] The handset 50 has a communication unit for communicating with the base unit via the network N, a call button 51 operated (pressed) by the person being monitored, a microphone 52, a speaker 53, etc. The base unit 40 has a lamp 41 provided in correspondence with each handset 50, a cancel button 42 for canceling the illumination of the lamp 41, a speaker 43, a handset 44, a control unit for controlling each of these units, etc. The lamps 41 and cancel buttons 42 are provided in the same number as the number of handset 50, and the handset 44 has a microphone and a speaker.

[0090] When the call button 51 of the handset 50 is operated by the person being monitored, the handset 50 transmits a call signal including information for identifying the handset itself (e.g., handset ID) to the base unit 40 via its communication unit. This notifies the base unit 40 of the information about the handset 50 whose call button 51 was operated. When the base unit 40 receives a call signal from any of the handset 50, it identifies the room number corresponding to the received handset ID, lights or flashes the lamp 41 of the identified room number, and outputs a call tone or call message from the speaker 43. This allows the room number of the handset 50 whose call button 51 was operated and the person being monitored to be notified to a care staff member near the base unit 40. When the care staff member picks up the receiver 44, a conversation (phone call) becomes possible between the handset 50 that sent the call signal and the base unit 40.

[0091] In this embodiment, the server 10 treats an operation on the call button 51 of the handset 50 as one of the events occurring to the monitored person. When the handset 50 calls the base unit 40, the server 10 stores the nurse call call as information related to the event in the event DB 13c. For example, the server 10 stores the current date and time, "call," and "nurse call" in the date and time column, event item column, and event content column of the event DB 13c. The server 10 also acquires the conversational audio exchanged between the base unit 40 and the handset 50, converts the acquired conversational audio into text data, and stores the conversational audio and conversational text in the conversational audio column and conversational text column of the event DB 13c, respectively. Therefore, in this embodiment, when a call is made using the handset 50, the text and audio related to the conversation exchanged between the monitored person and the caregiver via the handset 50 and the base unit 40 are stored in the event DB 13c.

[0092] The following describes the processing performed by the server 10 when a nurse call is made using the handset 50 in the information processing system of this embodiment. Fig. 12 is a flowchart showing an example of a processing procedure by the server 10 of embodiment 4. The following processing is executed by the control unit 11 in accordance with the program 13P stored in the auxiliary storage unit 13 of the server 10. The server 10 of this embodiment performs the following processing based on the sensor data sequentially received from the sensor I / F device 20 while performing the processing shown in Figs. 4 and 5 described in embodiment 1.

[0093] The control unit 11 of the server 10 determines whether a nurse call has been made by any of the handset 50 (S51). For example, the control unit 11 may be configured to be notified by the base unit 40 of the reception of a call signal when the base unit 40 receives a call signal from any of the handset 50. The control unit 11 may also be configured to be notified by the base unit 40 of the start of communication when the base unit 40 starts conversational voice communication with any of the handset 50. The control unit 11 can determine that a nurse call has been made based on such a notification from the base unit 40. If it is determined that a nurse call has not been made (S51: NO), the control unit 11 waits while performing other processing.

[0094] If it is determined that a nurse call has been made (S51: YES), the control unit 11 identifies the monitored person who made the call (S52). For example, the control unit 11 may obtain information about the monitored person who made the call through a notification from the base unit 40, or may identify the monitored person who made the call based on a call signal transmitted from the handset 50. The control unit 11 stores information about the event, that is, the nurse call call, in the event DB 13c of the identified monitored person (S53).

[0095] Next, the control unit 11 determines whether or not it has acquired voice data of the conversation transmitted and received between the handset 50 that made the nurse call and the base unit 40 (S54). The control unit 11 acquires, for example, voice data transmitted from the handset 50 to the base unit 40 via the network N, and voice data transmitted from the base unit 40 to the handset 50. Note that the control unit 11 may acquire the voice data transmitted and received between the handset 50 and the base unit 40 from the base unit 40 together with information indicating the speaker (the monitored person or the care staff member).

[0096] When the control unit 11 determines that voice data of the conversation has been acquired (S54: YES), it identifies the speaker of the acquired voice data (S55). For example, when the control unit 11 acquires voice data transmitted from the handset 50 to the base unit 40, it identifies the monitored person corresponding to the handset 50 (the monitored person identified in step S52) as the speaker of the voice data, and when the control unit 11 acquires voice data transmitted from the base unit 40 to the handset 50, it identifies the care staff member as the speaker of the voice data. Note that when a call from the handset 50 is answered using one of the staff terminals 30, the control unit 11 can identify the care staff member who is the user of the staff terminal 30 as the speaker of the voice data transmitted from the staff terminal 30 to the handset 50.

[0097] The control unit 11 converts the acquired voice data into text data and generates text data of the conversation between the care staff or the monitored person (S56). Then, the control unit 11 stores the text data generated in step S56 and the acquired voice data of the conversation in the event DB 13c of the monitored person identified in step S52 (S57). Specifically, the control unit 11 associates information indicating whether the speaker is the monitored person or the care staff, the date and time of reception of the voice data, and the generated text data of the conversation in a conversation text string, and associates information about the speaker, the date and time of reception of the voice data, and the acquired voice data in a conversation voice string. In this way, the control unit 11 can store the text data and voice data of the conversation between the monitored person and the care staff via the nurse call system in the event DB 13c of the monitored person.

[0098] When the control unit 11 determines that the voice data of the conversation has not been acquired (S54: NO), it skips the processes of steps S55 to S57. The control unit 11 determines whether the conversation via the base unit 40 and the handset 50 has ended (S58). For example, the control unit 11 determines that the conversation has ended when the transmission and reception of voice data between the base unit 40 and the handset 50 has ended. The control unit 11 may also be notified by the base unit 40 that the communication between the base unit 40 and the handset 50 has ended. When the control unit 11 determines that the conversation via the base unit 40 and the handset 50 has not ended (S58: NO), the control unit 11 returns to the process of step S54 and repeats the processes of steps S55 to S57 every time it acquires voice data of the conversation via the base unit 40 and the handset 50. When the control unit 11 determines that the conversation has ended (S58: YES), it ends the series of processes. At this time, the control unit 11 may store "Done" in the progress information column of the event DB 13c, indicating that the nursing care has ended.

[0099] Through the above-described processing, the server 10 can store text data and voice data of conversations between the monitored person and the care staff using the nurse call system. In this embodiment, too, the text data of the conversations stored in the server 10 can be viewed using the staff terminal 30 or another terminal, and a screen such as that shown in FIG. 7A can be generated. For example, in the screen shown in FIG. 7A, the top bold-line display frame displays information about the monitored person who made the nurse call call, the fact that the nurse call was made, and the date and time, and the second and subsequent display frames display text data of the conversation spoken by the care staff or monitored person together with the date and time of the call.

[0100] In this embodiment, the same effects as those of the above-described embodiments can be obtained. Furthermore, in this embodiment, text data and audio data of conversations between the person being monitored and the care staff can be stored using the nurse call system. Therefore, based on the stored data, it is possible to later check the response of the care staff when the person being monitored makes a nurse call call (the content of the nurse call response). The configuration of this embodiment can be applied to the information processing systems of the above-described embodiments 1 to 3, and similar effects can be obtained even when applied to the information processing systems of the above-described embodiments. Furthermore, in this embodiment, the modified examples described as appropriate in each of the above-described embodiments can also be applied.

[0101] (Embodiment 5) This description concerns an information processing system in which, when a notification event occurs to a monitored person, the server 10 extracts similar cases from the information stored in the event DB 13c and presents the nursing care provided in the extracted similar cases as advice to the nursing staff. The information processing system of this embodiment can be realized using devices similar to those of the information processing system of embodiment 1 shown in FIG. 1 , so a description of the configuration of each device will be omitted. In this embodiment, the monitored person DB 13a stores attribute information for each monitored person, including their age and gender, as well as medical records including their medical history, medical treatment history, medication history, current diagnosis and symptoms, and various test results such as blood tests and urine tests. The monitored person's medical records may be stored in the monitored person DB 13a or obtained from another server that stores electronic medical records including medical records.

[0102] Fig. 13 is a flowchart showing an example of a processing procedure in the information processing system of embodiment 5. The processing shown in Fig. 13 is obtained by adding steps S71 to S73 between YES in step S16 and step S17 in the processing shown in Fig. 4 and Fig. 5. Explanations of the same steps as in Fig. 4 and Fig. 5 will be omitted. Also, in Fig. 13, illustration of each step in Fig. 5 is omitted.

[0103] The server 10 and the staff terminal 30 of this embodiment perform the same processes as steps S11 to S16 in FIGS. 4 and 5. As a result, the server 10 detects the occurrence of an event for each monitored person based on signals from the sensors 21 to 25 and the call button 26 acquired via each sensor I / F device 20. If the control unit 11 of the server 10 determines that the event identified in step S14 is an event to be notified to the care staff (S16: YES), it extracts cases similar to the identified event from the contents of the event DB 13c for each monitored person. Specifically, the control unit 11 extracts monitored people whose attributes, symptoms, etc. are similar to those of the monitored person identified in step S12 from the attributes and medical records of each monitored person stored in the monitored person DB 13a (S71). For example, the control unit 11 extracts monitored people with the same diagnosis, monitored people with a similar type or progression of illness, monitored people taking the same medication, monitored people at the same stage of certification of needing support or long-term care, etc. The control unit 11 extracts a predetermined number (for example, five people) of monitoring targets who have more similar attributes and medical record contents.

[0104] Next, the control unit 11 extracts events (similar cases) similar to the event identified in step S14 from the information on each event stored in the event DB 13c of the extracted monitoring target (S72). For example, the control unit 11 extracts events with the same event item and event content, or events with the same event item and event content but occurring in a similar time period. The control unit 11 extracts a predetermined number of events (for example, one to two) that are more similar in event content, occurrence time period, etc.

[0105] The control unit 11 then generates advice based on the content of the care provided by the care staff in the extracted similar case (S73). For example, the control unit 11 reads out the conversation text stored in association with the extracted event (similar case) from the event DB 13c, and extracts phrases related to the content of the care provided by the care staff from the read conversation text. The control unit 11 then generates advice using the phrases related to the extracted content of the care. For example, in an event in which the monitored person with dementia gets out of bed and wanders outside the room, if the content of the care is extracted from the conversation text, such as "I said to him, 'Your family will come to pick you up soon, so please wait in your room,'" the control unit 11 generates the advice, such as "Please say to him, 'Your family will come to pick you up soon, so please wait in your room,'" if the content of the care is extracted from the conversation text, such as "He calmed down after taking a walk outside," the control unit 11 generates the advice, such as "Try taking a walk outside." The process of extracting the phrases related to the content of the care provided by the care staff from the conversation text stored in the event DB 13c may be performed in advance at an appropriate timing and stored in the event DB 13c. In this case, when generating advice in step S73, the control unit 11 can generate advice using the phrases extracted in advance, thereby simplifying the process.

[0106] The control unit 11 generates alert information to be notified to the care staff, including the advice generated in step S73 (S17). The alert information here includes the details of the event, information about the monitored person in whose presence the event occurred, and advice on the type of care the care staff should provide in response to the event. For example, in the case of an event in which a monitored person with dementia has left their bed and is wandering outside their room, the control unit 11 generates alert information including the monitored person's room number and name, a message indicating that the monitored person is wandering, and advice suggesting the type of care they should provide. Specifically, the control unit 11 generates alert information such as "Yamada Taro, Room 503, is wandering outside. Please take a walk outside." The control unit 11 and the control unit 31 of the staff terminal 30 then execute the processes from step S18 onward. Therefore, the staff terminal 30 can notify the care staff of the information about the monitored person in whose presence the event occurred, the details of the event, and the type of care the care staff should provide by audio output of the alert information received from the server 10. The alert information may be text data or image data, and if the staff terminal 30 has a display unit, the alert information may be displayed on the display unit to notify the care staff.

[0107] The above-described process not only notifies the care staff of the occurrence of a notification event in the monitored person, but also provides advice on the type of care the care staff can take based on similar past cases. In the above-described process, advice on the type of care the care staff can take is generated and presented to the care staff when the event occurs, but this configuration is not limited to this. For example, after the occurrence of an event in the monitored person is notified to the care staff, advice on the type of care the care staff can generate and present to the care staff while they are conversing using the staff terminal 30. In this case, advice on the type of care the care staff should take can be further provided based on the information about the monitored person and the type of event, as well as the type of care the care staff has provided up to that point.

[0108] In this embodiment, when extracting similar cases, the monitored person who is currently experiencing an event may be included in the monitored people with similar attributes and symptoms. In this case, advice can be presented based on the care provided by the care staff when a similar event occurred in the past for the same monitored person.

[0109] In this embodiment, when an event occurs to one of the monitored subjects, other monitored subjects with similar attributes and symptoms are identified from the information stored in the monitored subject DB 13a, and similar events are identified from the contents stored in the event DB 13c. For example, for a case (event) to be used for advice, the attributes and medical records of the monitored subject, the details of the event, and the details of care provided by the care staff are extracted in advance from the monitored subject DB 13a and the event DB 13c and stored in the advice DB. The server 10 may then identify monitored subjects with similar attributes and symptoms to the monitored subject in question and similar events from the contents stored in the advice DB, and acquire the details of care provided in association with the identified details. This configuration allows for quick search for similar cases to be used for advice, and by registering appropriate care details, appropriate advice can be provided. Furthermore, this advice DB is set up for each facility and periodically updated, allowing for customization based on the status of monitored subjects within the facility, thereby enabling the generation of advice tailored to each facility.

[0110] This embodiment achieves the same effects as the above-described embodiments. Furthermore, in this embodiment, when notifying a care staff member of an event occurring with respect to a monitored individual, the care staff member can be advised on the care care that the care staff member can take. This allows even care staff with limited skills, knowledge, or experience to provide appropriate care in accordance with the advice. Therefore, even when there are fewer care staff members, such as at night, it is possible to assign care staff with limited skills, knowledge, or experience. Furthermore, since care staff members can receive support through advice, anxiety about their work can be reduced, and a decrease in turnover and elimination of personnel shortages can be expected. The configuration of this embodiment can be applied to the information processing systems of the above-described embodiments 1 to 4, and similar effects can be achieved even when applied to the information processing systems of the above-described embodiments 1 to 4. Furthermore, the modified examples described in the above-described embodiments can also be applied to this embodiment.

[0111] The embodiments disclosed herein are illustrative in all respects and should not be considered limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0112] 10 Servers 11 Control section 12 Main memory 13 Auxiliary storage 14 Communications Department 20 Sensor I / F device 21 Sleep Sensor 22 Toilet Sensor 23 Door Sensor 24 Environmental Sensors 25 Human Sensor 26 Call button 30 Staff terminal 34 Communications Department 35. Mike 36 Earphones 13a Monitoring target database 13b Sensor data DB 13c Event DB

Claims

1. acquiring sensor data from a sensor that detects the state of a person to be monitored in a nursing facility or the state of the surroundings of the person to be monitored; Based on the acquired sensor data, an event occurring around the monitored person or the monitored person is determined; The occurrence of the determined event is output to the devices of the plurality of caregivers. Acquires voice data of conversations between the plurality of caregivers from the terminals of the plurality of caregivers; Convert the acquired voice data into text data, The system determines that the voice data of the conversation between the caregivers, acquired from the terminal of one caregiver within a predetermined time after outputting the occurrence of the event to the terminals of the plurality of caregivers, and the voice data of the conversation between the other caregivers, acquired from the terminal of the other caregiver within a predetermined time after acquiring the voice data, are conversations related to the event, and stores the determined event and text data converted from the voice data of the conversation between the plurality of caregivers regarding the event in association with each other in a storage unit. A program that causes a computer to perform a process.

2. It collects sensor data from multiple types of sensors, determining occurrence of a plurality of types of events based on the sensor data acquired from the plurality of types of sensors; An event of which occurrence should be output to the terminals of the plurality of caregivers is set for each of the monitoring subjects among the plurality of types of events. The program according to claim 1 , which causes the computer to execute a process.

3. The identification information of the caregiver and text data converted from the voice data acquired from the terminal of the caregiver are stored in the storage unit in association with each other.

3. The program according to claim 1, which causes the computer to execute processing.

4. extracting text data relating to the response made by the caregiver from the converted text data; The determined event and the extracted text data are stored in the storage unit in association with each other.

4. The program according to claim 1, which causes the computer to execute a process.

5. A report including information about the event is created based on the event and the text data stored in the storage unit.

5. The program according to claim 1, which causes the computer to execute a process.

6. The reports include a report for a care record, a report for reporting the event to a medical professional, and a report for reporting the event to a family member of the monitored person. The program according to claim 5.

7. Identifying keywords from the converted text data; Highlight the keyword in the text data 7. The program according to claim 1, which causes the computer to execute a process.

8. Detecting the end of the caregiver's response to the determined event; When the end of the care is detected, information indicating the end of the care by the caregiver is stored in the storage unit in association with the event.

8. The program according to claim 1, which causes the computer to execute a process.

9. Acquires voice data uttered by the person to be monitored, which is input via a nurse call handset provided in association with the person to be monitored; Acquire voice data uttered by the caregiver input via a nurse call master unit or a caregiver's terminal for responding to a call from the nurse call slave unit; converting the acquired voice data from the monitoring subject and the voice data from the caregiver into text data; Identification information of the person to be monitored is associated with text data converted from the voice data uttered by the person to be monitored, and identification information of the caregiver is associated with the text data converted from the voice data uttered by the caregiver. Based on the date and time of the event of a nurse call being made and the date and time of receipt of each voice data, the text data converted from the voice data of the conversation between the person to be monitored and the caregiver regarding the event of a nurse call being made is associated with the date and time of receipt of the voice data, and stored in a storage unit. A program that causes a computer to perform a process.

10. acquiring sensor data from a sensor that detects the state of a person to be monitored in a nursing facility or the state of the surroundings of the person to be monitored; Based on the acquired sensor data, an event occurring around the monitored person or the monitored person is determined; The occurrence of the determined event is output to the devices of the plurality of caregivers. Acquires voice data of conversations between the plurality of caregivers from the terminals of the plurality of caregivers; Convert the acquired voice data into text data, The system determines that the voice data of the conversation between the caregivers, acquired from the terminal of one caregiver within a predetermined time after outputting the occurrence of the event to the terminals of the plurality of caregivers, and the voice data of the conversation between the other caregivers, acquired from the terminal of the other caregiver within a predetermined time after acquiring the voice data, are conversations related to the event, and stores the determined event and text data converted from the voice data of the conversation between the plurality of caregivers regarding the event in association with each other in a storage unit. An information processing method in which processing is performed by a computer.

11. a sensor data acquisition unit that acquires sensor data from a sensor that detects the state of a person to be monitored in a nursing facility or the state of the surroundings of the person to be monitored; a determination unit that determines an event that has occurred around the person being monitored or the person being monitored based on the acquired sensor data; an output unit that outputs the occurrence of the determined event to a plurality of caregiver terminals; a voice data acquisition unit that acquires voice data of conversations between the plurality of caregivers from the plurality of caregiver terminals; a conversion unit that converts the acquired voice data into text data; a storage unit that determines that voice data of a conversation between the caregivers, acquired from the terminal of one caregiver within a predetermined time after outputting the occurrence of the event to the terminals of the plurality of caregivers, and voice data of conversations between the other caregivers, acquired from the terminals of the other caregivers within a predetermined time after acquiring the voice data, are conversations related to the event, and associates the determined event with text data converted from the voice data of the conversations between the plurality of caregivers regarding the event and stores the text data; An information processing device comprising:

Images