JP7603885B1 - Advice sending device and advice sending method - Google Patents

Abstract

The advice transmitting device (2) is configured to include a sensing data collecting unit (22) that collects sensing data from a sensor that monitors the condition of the care recipient, an advice determining unit (23) that determines advice to a care provider that provides care services to the care recipient based on the sensing data collected by the sensing data collecting unit (22), and an advice transmitting unit (24) that transmits the advice determined by the advice determining unit (23).

Description

The present disclosure relates to an advice sending device and an advice sending method.

There is an advice transmitting device that transmits advice to a user or the like.
As an example of such an advice transmitting device, Patent Document 1 discloses an advice transmitting device that generates advice and transmits the advice to a mobile terminal of a user.
The advice transmitting device acquires location information indicating the location of the mobile terminal from the user's mobile terminal. The advice transmitting device then generates advice by referring to a region table in which information on the region including the location indicated by the location information is recorded, and transmits the advice to the user's mobile terminal. The user is, for example, a person receiving care, such as an elderly person.

JP 2000-99442 A

The advice transmitting device disclosed in Patent Document 1 has a problem in that it cannot provide advice to the care provider when the condition of the care recipient, for example, the vital signs of the care recipient, change. The care provider is, for example, a staff member of a facility who provides care services to the care recipient.

This disclosure has been made to solve the problems described above, and aims to provide an advice transmitting device that can provide advice to a care provider when the condition of a care recipient changes.

The advice transmitting device according to the present disclosure includes a text data acquiring unit that acquires a notification request for advice to a care provider providing care services to a certain care recipient among one or more care recipients and outputs a command to collect sensing data related to the certain care recipient, a sensing data collecting unit that collects sensing data related to the certain care recipient from sensing data of sensors monitoring the status of each care recipient in accordance with the collection command output from the text data acquiring unit, an advice determining unit that determines advice to a care provider providing care services to the care recipient based on the sensing data collected by the sensing data collecting unit , and an advice transmitting unit that transmits the advice determined by the advice determining unit. The advice determining unit determines the necessity of advice based on the sensing data collected by the sensing data collecting unit, and if the necessity of advice exists, determines advice based on the sensing data.
The advice decision unit provides the sensing data collected by the sensing data collection unit to a first learning model realized by a neural network, obtains a judgment result of the need for advice from the first learning model, and if the judgment result of the need for advice indicates that there is a need for advice, provides the sensing data collected by the sensing data collection unit to a second learning model realized by a neural network, and obtains advice corresponding to the sensing data from the second learning model.
The first learning model is a model trained to output a judgment result on the need for advice in response to input of sensing data collected by the sensing data collection unit from the advice determination unit, and the second learning model is a model trained to output advice to a care provider corresponding to the sensing data in response to input of sensing data collected by the sensing data collection unit from the advice determination unit.

According to the present disclosure, advice can be provided to care providers when the situation of the person being cared for changes.

1 is a configuration diagram showing a system including an advice transmitting device 2 according to a first embodiment. 1 is a hardware configuration diagram showing hardware of an advice transmitting device 2 according to a first embodiment. FIG. 2 is a hardware configuration diagram of a computer in the case where the advice transmitting device 2 is realized by software, firmware, or the like. 4 is a flowchart showing an advice transmitting method which is a processing procedure of the advice transmitting device 2. FIG. 13 is an explanatory diagram showing an example of advice to staff who are care providers. FIG. 11 is a configuration diagram showing a system including an advice transmitting device 2 according to a second embodiment. FIG. 11 is a hardware configuration diagram showing hardware of an advice transmitting device 2 according to a second embodiment. FIG. 11 is a configuration diagram showing a system including an advice transmitting device 2 according to a third embodiment. FIG. 11 is a hardware configuration diagram showing hardware of an advice transmitting device 2 according to a third embodiment. 13 is a flowchart showing an example of a second advice determination performed by an advice determination unit 26; FIG. 13 is a configuration diagram showing a system including an advice transmitting device 2 according to a fourth embodiment. FIG. 13 is a hardware configuration diagram showing hardware of an advice transmitting device 2 according to a fourth embodiment.

In order to explain the present disclosure in more detail, the form for implementing the present disclosure will be described below with reference to the attached drawings.

Embodiment 1.
FIG. 1 is a configuration diagram showing a system including an advice transmitting device 2 according to the first embodiment.
FIG. 2 is a hardware configuration diagram showing the hardware of the advice transmitting device 2 according to the first embodiment.

The chatbot 1 comprises a voice recognition unit 11, a voice database 12, a chatbot engine unit 13, a display unit 14 and a voice output unit 15.
The chatbot 1 recognizes the user's voice and outputs text data indicating the user's voice to the advice sending device 2.
In the system shown in FIG. 1, the text data indicates a request for notification of advice, and the text data is output to the advice transmitting device 2 .
The chatbot 1 outputs the advice returned from the advice sending device 2 to both the display unit 14 and the audio output unit 15.

The voice recognition unit 11 recognizes the user's voice and creates text data representing the user's voice based on the voice data recorded in the voice database 12. The voice data recorded in the voice database 12 is, for example, voice piece data representing a voice piece.
The voice recognition unit 11 outputs the text data to the chatbot engine unit 13.
The speech database 12 stores, for example, speech unit data and text data corresponding to the speech unit data.

When the chatbot engine unit 13 obtains text data from the voice recognition unit 11, it outputs the text data to the advice sending device 2, and when text data is provided from a keyboard or the like (not shown), it outputs the text data to the advice sending device 2.
The chatbot engine unit 13 acquires advice from the advice transmitting device 2.
The chatbot engine unit 13 causes the display unit 14 to display the advice.
The chatbot engine unit 13 causes the audio output unit 15 to output the advice by voice.

The display unit 14 displays the advice output from the chatbot engine unit 13 on a display (not shown).
The audio output unit 15 outputs the advice output from the chatbot engine unit 13 as audio from a speaker (not shown).

The advice sending device 2 includes a text data acquiring unit 21 , a sensing data collecting unit 22 , an advice determining unit 23 , and an advice sending unit 24 .
The text data acquisition unit 21 is realized by, for example, a text data acquisition circuit 31 shown in FIG.
The text data acquisition unit 21 acquires text data indicating a request for notification of advice from the chatbot engine unit 13 of the chatbot 1 .
The text data acquisition unit 21 outputs the text data to the advice determination unit 23 .

The sensing data collection unit 22 is realized by, for example, a sensing data collection circuit 32 shown in FIG.
The sensing data collection unit 22 collects sensing data from various sensors included in the sensing device 3. The various sensors are sensors for monitoring the condition of the care recipient.
The condition of the care recipient may be, for example, the vital signs of the care recipient, the behavior of the care recipient, or the situation the care recipient is in. The condition of the care recipient may be, for example, the temperature, humidity, or surrounding noise in the care recipient's room.
The sensing data collection unit 22 outputs the sensing data to the advice determination unit 23 .

The advice determination unit 23 is realized by, for example, an advice determination circuit 33 shown in FIG.
When the advice determination unit 23 acquires text data indicating a request for notification of advice from the text data acquisition unit 21, it determines advice to a care provider who provides care services to the care recipient based on the sensing data collected by the sensing data collection unit 22.
The advice determining unit 23 outputs the determined advice to the advice transmitting unit 24 .

Specifically, the advice determination unit 23 provides the sensing data to the learning model 41 and obtains, from the learning model 41, advice corresponding to the sensing data.
The learning model 41 is realized by, for example, a neural network.
During learning, the learning model 41 is given sensing data related to the care recipient and advice to the care provider corresponding to the sensing data, and learns the advice to the care provider. The advice to the care provider is training data.
During inference, when sensing data is provided from the advice determination unit 23, the learning model 41 outputs advice corresponding to the sensing data.
1, the learning model 41 is provided outside the advice transmitting device 2. However, this is merely an example, and the learning model 41 may be provided inside the advice transmitting device 2.

The advice transmitting unit 24 is realized by, for example, an advice transmitting circuit 34 shown in FIG.
The advice sending unit 24 acquires advice to the care provider from the advice determining unit 23 .
The advice sending unit 24 sends the advice to the chatbot engine unit 13 of the chatbot 1.

1, it is assumed that each of the components of the advice sending device 2, that is, the text data acquisition unit 21, the sensing data collection unit 22, the advice determination unit 23, and the advice sending unit 24, is realized by dedicated hardware as shown in Fig. 2. In other words, it is assumed that the advice sending device 2 is realized by a text data acquisition circuit 31, a sensing data collection circuit 32, an advice determination circuit 33, and an advice sending circuit 34.
Each of the text data acquisition circuit 31, the sensing data collection circuit 32, the advice decision circuit 33 and the advice transmission circuit 34 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination of these.

The components of the advice transmitting device 2 are not limited to those realized by dedicated hardware, and the advice transmitting device 2 may be realized by software, firmware, or a combination of software and firmware.
The software or firmware is stored as a program in the memory of a computer. The computer means hardware that executes the program, and includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor).

FIG. 3 is a hardware configuration diagram of a computer in the case where the advice transmitting device 2 is realized by software, firmware, or the like.
When the advice sending device 2 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures of the text data acquiring unit 21, the sensing data collecting unit 22, the advice determining unit 23, and the advice sending unit 24 are stored in the memory 51. Then, a processor 52 of the computer executes the programs stored in the memory 51.

Furthermore, Fig. 2 shows an example in which each of the components of the advice transmitting device 2 is realized by dedicated hardware, and Fig. 3 shows an example in which the advice transmitting device 2 is realized by software or firmware, etc. However, this is merely one example, and some of the components in the advice transmitting device 2 may be realized by dedicated hardware, and the remaining components may be realized by software or firmware, etc.

The sensing device 3 is equipped with sensors for monitoring the condition of the care recipient, such as a bed sensor 61, a room sensor 62, an appliance sensor 63, a human presence sensor 64, a vital sign sensor 65 and an intake detection sensor 66, and the sensing device 3 is also equipped with a sensing data recording unit 67.
The bed sensor 61 is a sensor attached to the bed used by the care recipient, and monitors, for example, the state of the care recipient turning over in his/her sleep.
The room sensor 62 is a sensor installed in the room of the care recipient, and monitors, for example, the sleep, falls, or body movements of the care recipient.

The home appliance sensor 63 is, for example, a sensor attached to an air conditioner in the care recipient's room, and monitors, for example, the room temperature in the care recipient's room.
The human presence sensor 64 is, for example, a sensor attached to the entrance of the care recipient's room, and detects, for example, when the care recipient goes out.
The vital sign sensor 65 is, for example, a sensor attached to a wall, ceiling, or furniture in the care recipient's room, or to the care recipient, and monitors, for example, the vital signs of the care recipient.
The ingestion detection sensor 66 is, for example, a sensor including a camera attached to the wall of the care recipient's room, and monitors, for example, the fluid intake behavior or food intake behavior of the care recipient.
The sensing data recording unit 67 is a database that records the sensing data from the bed sensor 61, the room sensor 62, the home appliance sensor 63, the human presence sensor 64, the vital sensor 65, and the ingestion detection sensor 66.

Next, the operation of the advice transmitting device 2 shown in FIG. 1 will be described.
FIG. 4 is a flowchart showing an advice transmitting method which is a processing procedure of the advice transmitting device 2.
In the system shown in FIG. 1, the care recipient is, for example, a person admitted to a nursing home or a medical hospital, and the care provider is, for example, a staff member of the nursing home or the medical hospital.

When the voice recognition unit 11 of the chatbot 1 receives, for example, the voice of a staff member from a microphone not shown, it breaks down the voice of the staff member into, for example, multiple sound fragments.
The speech recognition unit 11 obtains text data corresponding to each speech piece from the speech database 12 .
The voice recognition unit 11 connects together the text data corresponding to each speech piece and outputs the connected text data to the chatbot engine unit 13.
Examples of text data include a notification request for advice such as "Please give me some advice on providing nursing care services to Mr. X", or a notification request for advice such as "Please give me some advice on nursing care services".

Here, the staff member's voice is provided to the voice recognition unit 11, and text data is output from the voice recognition unit 11 to the chatbot engine unit 13. However, this is merely one example, and the staff member may operate a keyboard or the like (not shown) to provide text data to the chatbot engine unit 13 from the keyboard or the like.

The chatbot engine unit 13 acquires text data indicating a request for notification of advice from the voice recognition unit 11.
The chatbot engine unit 13 outputs the text data to the text data acquisition unit 21 of the advice sending device 2.

The text data acquisition unit 21 acquires text data indicating a request for notification of advice from the chatbot engine unit 13 of the chatbot 1 (step ST1 in FIG. 4).
The text data acquisition unit 21 outputs the text data to the advice determination unit 23 .
The text data acquisition unit 21 also identifies the care recipient by performing a process for identifying the care recipient. The process for identifying the care recipient may be, for example, a known process for identifying who the care recipient is by searching for the name of the care recipient contained in the text data.
If the advice notification request is for advice on nursing care services for Mr./Ms. XX among one or more care recipients, the text data acquisition unit 21 outputs a collection command for sensing data related to Mr./Ms. XX to the sensing data collection unit 22. If the advice notification request is for advice on nursing care services for Mr./Ms. △△ among one or more care recipients, the text data acquisition unit 21 outputs a collection command for sensing data related to Mr./Ms. △△ to the sensing data collection unit 22.
If the care recipient is not specified, the text data acquiring unit 21 does not limit the sensing data and outputs a command to collect sensing data to the sensing data collecting unit 22 .

When the sensing data collection unit 22 receives an instruction to collect sensing data related to the care recipient from the text data acquisition unit 21, it collects the sensing data related to the care recipient from the sensing data recorded in the sensing data recording unit 67 of the sensing device 3 (step ST2 in Figure 4).
The sensing data collection unit 22 collects sensing data relating to a care recipient named XX if the care recipient is XX, and collects sensing data relating to XX if the care recipient is △△.
If the care recipient is not specified, the sensing data collection unit 22 collects all the sensing data recorded in the sensing data recording unit 67 .
The sensing data collection unit 22 outputs the sensing data to the advice determination unit 23 .

When the advice determination unit 23 acquires text data indicating a request for notification of advice from the text data acquisition unit 21, it determines advice to a care provider who provides care services to the care recipient based on the sensing data collected by the sensing data collection unit 22 (step ST3 in Figure 4).
The advice determining unit 23 outputs the determined advice to the advice transmitting unit 24 .
The advice determination process performed by the advice determination unit 23 will now be described in detail.

The advice determination unit 23 provides the sensing data to the learning model 41 and obtains, from the learning model 41, advice corresponding to the sensing data.
If the sensing data collected by the sensing data collection unit 22 is, for example, sensing data from a vital sensor 65 and the sensing data is, for example, Mr./Ms. XX's body temperature and indicates that Mr./Ms. XX's body temperature is higher than normal, then the learning model 41 outputs advice such as those shown in (1) to (5) below.
(1) Please check Mr./Ms. XX’s hydration level.
(2) Check the room temperature.
(3) You suspect you have a urinary tract infection.
(4) Please cool down Mr./Ms. XX so that his/her fever can go down.
(5) Check for chills.

If the sensing data collected by the sensing data collection unit 22 is, for example, from an intake detection sensor 66, and the sensing data indicates, for example, that Mr./Ms. X's food intake behavior is declining, the learning model 41 outputs advice such as those shown in (1) to (4) below.
(1) Mr./Ms. XX's appetite may be decreasing.
(2) Please check Mr./Ms. XX’s medication status.
(3) Please check Mr./Ms. XX’s vital signs.
(4) Check whether or not you have snacks.

The sensing data collected by the sensing data collection unit 22 is, for example, a bed sensor 61. If the sensing data indicates, for example, that Mr./Ms. X's sleep behavior is declining, the learning model 41 outputs advice such as those shown in (1) to (4) below.
(1) Mr./Ms. XX may not be getting enough sleep.
(2) Please check the amount of sleeping pills that Mr./Ms. XX is taking.
(3) Ask Mr./Ms. XX if he/she has any worries.
(4) Ask Mr./Ms. XX if there is any part of his/her body that is painful.

The sensing data collected by the sensing data collection unit 22 is, for example, a room sensor 62. If the sensing data indicates, for example, an abnormality in Mr./Ms. X's physical movement behavior, the learning model 41 outputs advice such as those shown in (1) to (3) below.
(1) Please check Mr./Ms. XX's cognitive function.
(2) Ask Mr./Ms. XX if he/she has any worries.
(3) Please check Mr./Ms. XX’s medication status.

The advice sending unit 24 acquires advice to the care provider from the advice determining unit 23 .
The advice sending unit 24 sends the advice to the chatbot engine unit 13 of the chatbot 1 (step ST4 in FIG. 4).
When multiple pieces of advice are output from the advice determination unit 23, the advice sending unit 24 may send all of the multiple pieces of advice to the chatbot engine unit 13, or may send one or more pieces of advice to the chatbot engine unit 13.

The chatbot engine unit 13 obtains advice from the advice sending unit 24.
The chatbot engine unit 13 displays the advice on the display of the display unit 14, as shown in FIG.
In addition, the chatbot engine unit 13 outputs the advice as voice from the speaker of the voice output unit 15.
FIG. 5 is an explanatory diagram showing an example of advice given to staff who are care providers.
In FIG. 5, □□□□ is the name of a staff member.
In the example of FIG. 5, the advice "Check whether you have chills or not" is displayed.

In the above-described first embodiment, the advice transmitting device 2 is configured to include a sensing data collecting unit 22 that collects sensing data from a sensor that monitors the condition of the care recipient, an advice determining unit 23 that determines advice to a care provider who provides care services to the care recipient based on the sensing data collected by the sensing data collecting unit 22, and an advice transmitting unit 24 that transmits the advice determined by the advice determining unit 23. Thus, the advice transmitting device 2 can provide advice to the care provider when the condition of the care recipient changes.

In the advice sending device 2 shown in Fig. 1, when the text data acquisition unit 21 acquires text data indicating a request for notification of advice from the chatbot 1, the advice determination unit 23 determines advice for the care provider, and the advice sending unit 24 sends the advice to the chatbot 1. However, this is merely one example, and the advice determination unit 23 may determine advice for the care provider and the advice sending unit 24 may send the advice to the chatbot 1, regardless of whether the text data acquisition unit 21 acquires text data indicating a request for notification of advice from the chatbot 1.

In the advice sending device 2 shown in Fig. 1, the advice determination unit 23 provides sensing data to the learning model 41 and obtains advice corresponding to the sensing data from the learning model 41. However, this is merely one example, and the advice determination unit 23 may determine advice corresponding to the sensing data based on a rule base, for example.

In the advice transmitting device 2 shown in Fig. 1, the text data indicating the advice notification request includes information indicating that the care recipient is, for example, "Mr. XX" or "Mr. △△". However, this is merely an example, and the text data does not necessarily include information indicating that the care recipient is, for example, "Mr. XX" or "Mr. △△". In this case, for example, when a staff member at a nursing home or the like gives a voice to the voice recognition unit 11 of the chatbot 1, the staff member's ID (IDentification) is provided to the chatbot engine unit 13, and the chatbot engine unit 13 displays information indicating one or more care recipients that the staff member is in charge of on the display unit 14.
Then, the chatbot engine unit 13 accepts a selection operation by a staff member to select a care recipient related to the advice notification request from one or more care recipients, and outputs care recipient information indicating the care recipient related to the advice notification request to the advice sending device 2. The text data acquisition unit 21 outputs a collection command for sensing data related to the care recipient indicated by the care recipient information to the sensing data collection unit 22.

Embodiment 2.
In the second embodiment, an advice transmitting device 2 in which an advice determining unit 25 determines the necessity of advice based on sensing data collected by a sensing data collecting unit 22 will be described.

Fig. 6 is a configuration diagram showing a system including an advice transmitting device 2 according to embodiment 2. In Fig. 6, the same reference numerals as in Fig. 1 indicate the same or corresponding parts, and detailed description thereof will be omitted.
Fig. 7 is a hardware configuration diagram showing the hardware of the advice transmitting device 2 according to embodiment 2. In Fig. 7, the same reference numerals as in Fig. 2 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted.
The advice sending device 2 includes a text data acquiring unit 21 , a sensing data collecting unit 22 , an advice determining unit 25 , and an advice sending unit 24 .

The advice determination unit 25 is realized by, for example, an advice determination circuit 35 shown in FIG.
When the advice determination section 25 acquires text data indicating a request for notification of advice from the text data acquisition section 21 , the advice determination section 25 determines the necessity of advice based on the sensing data collected by the sensing data collection section 22 .
If advice is necessary, the advice determining unit 25 determines advice to the care provider based on the sensing data collected by the sensing data collecting unit 22 .
The advice determining unit 25 outputs the determined advice to the advice transmitting unit 24 .

Specifically, the advice determination unit 25 provides the sensing data to the learning model 42 and acquires, from the learning model 42, a determination result as to the necessity of advice.
If the necessity determination result indicates that advice is necessary, the advice determination unit 25 provides the sensing data to the learning model 41 and obtains, from the learning model 41, advice corresponding to the sensing data.
The learning model 42 is realized by, for example, a neural network.
During learning, the learning model 42 is given the sensing data related to the care recipient and information indicating whether or not advice is necessary, and learns whether or not advice is necessary. The information indicating whether or not advice is necessary is teacher data.
During inference, when sensing data is provided from the advice determination unit 23, the learning model 42 outputs a determination result of the necessity of advice.
6, the learning model 42 is provided outside the advice transmitting device 2. However, this is merely an example, and the learning model 42 may be provided inside the advice transmitting device 2.

In Fig. 6, it is assumed that each of the components of the advice sending device 2, that is, the text data acquiring unit 21, the sensing data collecting unit 22, the advice determining unit 25 and the advice sending unit 24, is realized by dedicated hardware as shown in Fig. 7. In other words, it is assumed that the advice sending device 2 is realized by a text data acquiring circuit 31, a sensing data collecting circuit 32, an advice determining circuit 35 and an advice sending circuit 34.
Each of the text data acquisition circuit 31, the sensing data collection circuit 32, the advice decision circuit 35 and the advice transmission circuit 34 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these.

The components of the advice transmitting device 2 are not limited to those realized by dedicated hardware, and the advice transmitting device 2 may be realized by software, firmware, or a combination of software and firmware.
When the advice transmitting device 2 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures of the text data acquiring unit 21, the sensing data collecting unit 22, the advice determining unit 25, and the advice transmitting unit 24 are stored in a memory 51 shown in Fig. 3. Then, a processor 52 shown in Fig. 3 executes the programs stored in the memory 51.

Furthermore, Fig. 7 shows an example in which each of the components of the advice transmitting device 2 is realized by dedicated hardware, and Fig. 3 shows an example in which the advice transmitting device 2 is realized by software or firmware, etc. However, this is merely one example, and some of the components in the advice transmitting device 2 may be realized by dedicated hardware, and the remaining components may be realized by software or firmware, etc.

Next, the operation of the advice transmitting device 2 shown in Figure 6 will be described. However, apart from the advice determination unit 25, the advice transmitting device 2 is the same as that shown in Figure 1. Therefore, here, the operation of the advice determination unit 25 will mainly be described.

When the advice determination unit 25 acquires text data indicating a request for notification of advice from the text data acquisition unit 21 , it acquires sensing data related to the care recipient from the sensing data collection unit 22 .
The advice determination unit 25 provides sensing data relating to the care recipient to the learning model 42 and acquires, from the learning model 42, a determination result as to the necessity of advice.
If the judgment result output from the learning model 42 indicates that advice is necessary, the advice determination unit 25 provides the sensing data to the learning model 41 and obtains advice from the learning model 41.
The advice determining unit 25 outputs the acquired advice to the advice transmitting unit 24 .
If the judgment result output from the learning model 42 indicates that there is no need for advice, the advice determination unit 25 does not perform the advice determination process.

In the above-described second embodiment, the advice transmitting device 2 is configured so that the advice determining unit 25 determines the need for advice based on the sensing data collected by the sensing data collecting unit 22, and if there is a need for advice, determines the advice based on the sensing data. Therefore, even if the situation of the care recipient changes, the advice transmitting device 2 can provide advice to the care provider only when the need for advice is recognized.

In the advice sending device 2 shown in Fig. 6, the advice determination unit 25 provides sensing data related to the care recipient to the learning model 42 and obtains a determination result of the need for advice from the learning model 42. However, this is merely one example, and the advice determination unit 25 may determine the need for advice based on a rule base, for example.

Embodiment 3.
In embodiment 3, an advice sending device 2 is described in which the advice determination unit 26 determines the need for advice based not only on the sensing data collected by the sensing data collection unit but also on the care record of the care recipient or the medication information of the care recipient.

Fig. 8 is a configuration diagram showing a system including an advice transmitting device 2 according to embodiment 3. In Fig. 8, the same reference numerals as in Fig. 1 and Fig. 6 indicate the same or corresponding parts, and detailed description thereof will be omitted.
Fig. 9 is a hardware configuration diagram showing the hardware of the advice transmitting device 2 according to embodiment 3. In Fig. 9, the same reference numerals as in Fig. 2 and Fig. 7 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted.
The advice sending device 2 includes a text data acquiring unit 21 , a sensing data collecting unit 22 , an advice determining unit 26 , and an advice sending unit 24 .

The advice determination unit 26 is realized by, for example, an advice determination circuit 36 shown in FIG.
When the advice determination unit 26 acquires text data indicating a request for notification of advice from the text data acquisition unit 21, it determines the need for advice based on the sensing data collected by the sensing data collection unit 22 and the care records of the care recipient stored in the care record database 45.
If advice is necessary, the advice determining unit 26 determines advice to the care provider based on the sensing data collected by the sensing data collecting unit 22 and the care record of the care recipient.
The advice determining unit 26 outputs the determined advice to the advice transmitting unit 24 .

Specifically, the advice determination unit 26 provides the sensing data and the care record of the care recipient to the learning model 44 and obtains, from the learning model 44, a determination result of the necessity of advice.
If the necessity judgment result indicates that advice is necessary, the advice determination unit 26 provides the sensing data and the care record of the care recipient to the learning model 43, and obtains advice corresponding to both the sensing data and the care record from the learning model 43.

The learning model 43 is realized by, for example, a neural network.
During learning, the learning model 43 is given sensing data related to the care recipient, a care record of the care recipient, and advice to the care provider corresponding to both the sensing data and the care record, and learns advice to the care provider. The advice to the care provider is teacher data.
During inference, when the learning model 43 receives the sensing data and the care record of the care recipient from the advice determination unit 26, the learning model 43 outputs advice corresponding to both the sensing data and the care record.

The learning model 44 is realized by, for example, a neural network.
During learning, the learning model 44 is given sensing data related to the care recipient, the care record of the care recipient, and information indicating whether or not advice is necessary, and learns whether or not advice is necessary. The information indicating whether or not advice is necessary is teacher data.
During inference, when the learning model 44 receives the sensing data and the care record of the care recipient from the advice determination unit 26, the learning model 44 outputs a determination result of the necessity of advice.
8, the learning models 43, 44 are provided outside the advice transmitting device 2. However, this is merely an example, and the learning models 43, 44 may be provided inside the advice transmitting device 2.

Here, the advice determination unit 26 determines the necessity of advice based on the sensing data and the care record of the care recipient. However, this is merely an example, and the advice determination unit 26 may determine the necessity of advice based on the sensing data and the medication information of the care recipient stored in the care record database 45. In this case, the advice determination unit 26 determines advice to the care provider based on the sensing data and the medication information of the care recipient.
The advice determination unit 26 may also determine the necessity of advice based on the sensing data, the care record of the care recipient, and the medication information of the care recipient. In this case, the advice determination unit 26 determines advice to the care provider based on the sensing data, the care record of the care recipient, and the medication information of the care recipient.
The advice determination unit 26 may also determine the necessity of advice based on the sensing data, the care record of the care recipient, the medication information of the care recipient, medical information described below, and care information described below.

The care record database 45 stores the care records of the care recipients.
The care record database 45 also stores, as the care record of the care recipient, information indicating the medical history of the care recipient, information of the care recipient recorded on a face sheet, or the schedule of the care recipient. The information of the care recipient recorded on the face sheet includes, for example, the name, age, address, family structure, hobbies, preferences, or work history of the care recipient. The schedule of the care recipient includes, for example, a schedule for visiting the hospital, a schedule for attending hobby classes, or a schedule for going to the bank. The care record database 45 also stores the medical visit history of the care recipient, as well as medication information of the care recipient. The medication information is information indicating which care recipient took which medicine and at what time.
The medical and nursing care information database 46 stores medical information and nursing care information. The medical information is information contained in general medical textbooks, such as symptoms, medicines, or medical treatments. The nursing care information is information contained in textbooks for nursing care workers, such as care related to nursing care or nursing care treatments.
1 , the learning models 43, 44, the care record database 45, and the medical and nursing care information database 46 are provided outside the advice transmitting device 2. However, this is merely an example, and the learning models 43, 44, the care record database 45, and the medical and nursing care information database 46 may be provided inside the advice transmitting device 2.

In Fig. 8, it is assumed that each of the components of the advice sending device 2, that is, the text data acquiring unit 21, the sensing data collecting unit 22, the advice determining unit 26, and the advice sending unit 24, is realized by dedicated hardware as shown in Fig. 9. In other words, it is assumed that the advice sending device 2 is realized by a text data acquiring circuit 31, a sensing data collecting circuit 32, an advice determining circuit 36, and an advice sending circuit 34.
Each of the text data acquisition circuit 31, the sensing data collection circuit 32, the advice decision circuit 36 and the advice transmission circuit 34 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these.

The components of the advice transmitting device 2 are not limited to those realized by dedicated hardware, and the advice transmitting device 2 may be realized by software, firmware, or a combination of software and firmware.
When the advice transmitting device 2 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures in the text data acquiring unit 21, the sensing data collecting unit 22, the advice determining unit 26, and the advice transmitting unit 24 are stored in a memory 51 shown in Fig. 3. Then, a processor 52 shown in Fig. 3 executes the programs stored in the memory 51.

Furthermore, Fig. 9 shows an example in which each of the components of the advice transmitting device 2 is realized by dedicated hardware, and Fig. 3 shows an example in which the advice transmitting device 2 is realized by software or firmware, etc. However, this is merely one example, and some of the components in the advice transmitting device 2 may be realized by dedicated hardware, and the remaining components may be realized by software or firmware, etc.

Next, the operation of the advice transmitting device 2 shown in Figure 8 will be described. However, apart from the advice determination unit 26, the advice transmitting device 2 is the same as that shown in Figure 1. Therefore, here, the operation of the advice determination unit 26 will mainly be described.

When the advice determining unit 26 acquires text data indicating a request for notification of advice from the text data acquiring unit 21 , it acquires sensing data related to the care recipient from the sensing data collecting unit 22 .
In addition, the advice determining unit 26 acquires the care record of the care recipient and the medication information of the care recipient from the care record database 45 .
Here, the advice determination unit 26 acquires the care record of the care recipient and the medication information of the care recipient. However, this is merely an example, and the advice determination unit 26 may acquire either the care record of the care recipient or the medication information of the care recipient.

The advice determination unit 26 provides the sensing data related to the care recipient, the care record of the care recipient, and the medication information of the care recipient to the learning model 44, and obtains from the learning model 44 a determination result of the necessity of advice.
The sensing data relating to the care recipient is, for example, sensing data from the bed sensor 61, and the sensing data indicates, for example, that Mr./Ms. XX is not sleeping well. In this case, if mental symptoms are recorded in Mr./Ms. XX's care record, or if medication for depression is recorded in Mr./Ms. XX's medication information, the learning model 44 outputs a determination result that advice is necessary.
On the other hand, if no mental symptoms are recorded in Mr. X's nursing care records and no medication for depression is recorded in Mr. X's medication information, the learning model 44 will output a determination result that no advice is necessary.

Assume that the sensing data related to the care recipient is, for example, from the room sensor 62, and the sensing data indicates, for example, that Mr./Ms. XX is moving around. At this time, if there is a description about dementia recorded in Mr./Ms. XX's care record, or if medicine for dementia is recorded in Mr./Ms. XX's medication information, the learning model 44 outputs a determination result that advice is necessary.
On the other hand, if there is no mention of dementia in Mr. X's nursing care records and no medication related to dementia is recorded in Mr. X's medication information, the learning model 44 will output a determination result that no advice is necessary.

If the judgment result output from the learning model 44 indicates that advice is necessary, the advice determination unit 26 provides the sensing data, the care record of the care recipient, and the medication information of the care recipient to the learning model 43, and obtains advice from the learning model 43.
When the sensing data indicates, for example, that Mr./Ms. X is not sleeping, when mental symptoms are recorded in Mr./Ms. X's nursing records, or when medication for depression is recorded in Mr./Ms. X's medication information, the learning model 43 outputs advice such as those shown in (1) to (3) below.
(1) Ask Mr./Ms. XX if he/she has any worries.
(2) Ask Mr./Ms. XX if he/she is staying hydrated.
(3) Please ask Mr./Ms. XX if he/she is taking any medication.

When the sensing data indicates, for example, that Mr./Ms. X is moving around, when there is a description of dementia recorded in Mr./Ms. X's nursing care records, or when medication for dementia is recorded in Mr./Ms. X's medication information, the learning model 43 outputs advice such as those shown in (1) to (3) below.
(1) Please check Mr./Ms. XX's cognitive function.
(2) Ask Mr./Ms. XX if he/she has any worries.
(3) Please check Mr./Ms. XX’s medication status.

The advice determining unit 26 outputs the acquired advice to the advice transmitting unit 24 .
If the judgment result output from the learning model 43 indicates that there is no need for advice, the advice determination unit 26 does not perform the advice determination process.

In the above-described third embodiment, the advice transmitting device 2 shown in FIG. 8 is configured so that the advice determining unit 26 determines the need for advice based on the sensing data collected by the sensing data collecting unit 22 and the care record of the care recipient, and if there is a need for advice, determines the advice based on the sensing data and the care record. Therefore, the advice transmitting device 2 shown in FIG. 8 can provide more appropriate advice than the advice transmitting device 2 shown in FIG. 1.

In addition, in embodiment 3, the advice transmitting device 2 shown in Fig. 8 is configured so that the advice determining unit 26 determines the need for advice based on the sensing data collected by the sensing data collecting unit 22 and the medication information of the care recipient, and if there is a need for advice, determines the advice based on the sensing data and the medication information. Therefore, the advice transmitting device 2 shown in Fig. 8 can provide more appropriate advice than the advice transmitting device 2 shown in Fig. 1.

In the advice transmitting device 2 according to the first to third embodiments, the advice determination units 23, 25, 26 provide one type of sensing data to the learning models 41, 43, and acquire advice corresponding to the sensing data from the learning models 41, 43. However, this is merely an example, and the advice determination units 23 etc. may provide a plurality of sensing data to the learning models 41 etc., and acquire advice corresponding to the plurality of sensing data from the learning models 41 etc.
Below are listed specific examples showing the relationship between abnormalities based on sensing data, nursing care records, medication information, and advice.

(1) Specific example (1)
・Anomaly based on sensing data → Anomaly (1) + Anomaly (2)
Abnormality (1) = Not being able to sleep at all: An abnormality detected from sensing data of the bed sensor 61, etc. Abnormality (2) = Fluctuations in blood pressure: An abnormality detected from sensing data of the vital sensor 65. Nursing records → There is a description of mental symptoms.
- Medication information: There is information about taking tranquilizers.
・Advice → Ask them if they are worried about anything in their daily life.
Check to see if you have a fever.
Make sure you are staying hydrated.
Make sure you are taking your medication.

(2) Specific example (2)
・Anomaly based on sensing data → Anomaly (1) + Anomaly (2)
Abnormality (1) = high body temperature: abnormality detected from sensing data of vital sensor 65. Abnormality (2) = low fluid intake: abnormality detected from sensing data of intake detection sensor 66. Nursing records: description of cold symptoms, description of urinary tract infection, description of pneumonia, description of reluctance to drink fluids, or description of sleeping in thick clothes.
-Medication information → None in particular -Advice → Please check the amount of fluid you are drinking.
Check the room temperature.
You suspect a urinary tract infection.
Cool the area to reduce the fever.

(3) Specific example (3)
・Anomaly based on sensing data → Anomaly (1) + Anomaly (2) + Anomaly (3)
Abnormality (1) = Restlessness: An abnormality detected from the sensing data of the room sensor 62. Abnormality (2) = Small amount of food eaten: An abnormality detected from the sensing data of the intake detection sensor 66. Abnormality (3) = Leaving the room multiple times: An abnormality detected from the sensing data of the human presence sensor 64. Nursing care records: There are descriptions of dementia or cold symptoms.
- Medication information → None in particular - Advice → Please check for decline in cognitive function.
Please consult your supervisor.
Please speak up.
Check your medication history.

(4) Specific example (4)
・Anomaly based on sensing data → Anomaly (1) + Anomaly (2) + Anomaly (3)
Abnormality (1) = Decreased amount of food eaten: abnormality detected from sensing data of intake detection sensor 66. Abnormality (2) = Decreased amount of water consumed: abnormality detected from sensing data of intake detection sensor 66. Abnormality (3) = Loss of skin firmness: abnormality detected from sensing data of vital sensor 65. Nursing records: description of elderly age or description of malnutrition.
-Medication information → None in particular -Advice → Check your hydration status.
Please consult your doctor.

As described above, by configuring the advice sending device 2 so that the advice determination unit 23 etc. determines advice to the care provider based on multiple sensing data collected by the sensing data collection unit 22, it is possible to optimize the advice when multiple abnormalities occur.

In the third embodiment, the advice transmitting device 2 shown in Fig. 8 is configured so that the advice determining unit 26 determines advice based on sensing data, etc., if advice is necessary. When the care provider who has confirmed the advice acts in accordance with the advice and text data indicating the result of the action is provided to the text data acquiring unit 21 (step ST11 in Fig. 10: YES), the advice determining unit 26 may further determine advice for the result of the action (steps ST12 to ST14 in Fig. 10).
Hereinafter, when the advice determined by the advice determination unit 26 is, for example, "Check if you have a fever," an example of further advice determined will be specifically described.
FIG. 10 is a flowchart showing an example of how the advice determination unit 26 determines a second piece of advice.
Here, it is assumed that after the advice determination unit 26 determines the advice "Please check if the person has a fever," the care provider follows the advice, takes the temperature of the person being cared for, and text data indicating the body temperature of the person being cared for is provided to the text data acquisition unit 21.
In such a case, if the body temperature of the care recipient is equal to or higher than the threshold value (step ST12 in FIG. 10: YES), the advice determination unit 26 determines advice for when the body temperature of the care recipient is high (step ST13 in FIG. 10). Examples of such advice include advice such as "Please have the care recipient examined as a specific illness is suspected" or advice such as "Please have the care recipient replenish with fluids as there is a possibility of dehydration." The threshold value may be stored in the internal memory of the advice determination unit 26, or may be provided from outside the advice transmission device 2.
If the body temperature of the care recipient is below the threshold (step ST12 in FIG. 10: NO), the advice determination unit 26 determines advice for when the body temperature of the care recipient is not high (step ST14 in FIG. 10). For example, such advice is "Observe the condition and check the temperature again in one hour."
The advice determining unit 26 outputs the determined advice to the advice transmitting unit 24 .

Embodiment 4.
In the fourth embodiment, an advice transmitting device 2 in which the advice transmitting unit 27 determines the timing of transmitting advice based on the advice determined by the advice determining unit 26 will be described.

Fig. 11 is a configuration diagram showing a system including an advice transmitting device 2 according to embodiment 4. In Fig. 11, the same reference numerals as those in Fig. 1, Fig. 6 and Fig. 8 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted.
Fig. 12 is a hardware configuration diagram showing the hardware of the advice transmitting device 2 according to embodiment 4. In Fig. 12, the same reference numerals as those in Fig. 2, Fig. 7, and Fig. 9 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted.
The advice sending device 2 includes a text data acquiring unit 21 , a sensing data collecting unit 22 , an advice determining unit 23 , and an advice sending unit 27 .

The advice transmitting unit 27 is realized by, for example, an advice transmitting circuit 37 shown in FIG.
The advice sending unit 27 acquires advice to the care provider from the advice determining unit 23 .
The advice transmitting unit 27 determines the timing for transmitting the advice based on the advice determined by the advice determining unit 26 .
Specifically, the advice sending unit 27 gives the advice determined by the advice determining unit 26 to the learning model 47 and obtains, from the learning model 47, information indicating the advice transmission timing.
The advice sending unit 27 sends the advice to the chatbot engine unit 13 of the chatbot 1 at the determined timing.

The learning model 47 is realized by, for example, a neural network.
During learning, the learning model 47 is given advice and information indicating the advice transmission timing, and learns the advice transmission timing. The information indicating the advice transmission timing is teacher data.
During inference, when advice is given from the advice sending unit 27, the learning model 47 outputs information indicating the advice sending timing.
11, the learning model 47 is provided outside the advice transmitting device 2. However, this is merely an example, and the learning model 47 may be provided inside the advice transmitting device 2.

In the system shown in Figure 11, the advice sending unit 27 and the learning model 47 are each applied to the system shown in Figure 1. However, this is merely an example, and the advice sending unit 27 and the learning model 47 may each be applied to the system shown in Figure 6 or the system shown in Figure 8.

Next, the operation of the advice transmitting device 2 shown in Figure 11 will be described. However, apart from the advice transmitting unit 27, the advice transmitting device 2 is the same as that shown in Figure 1. Therefore, only the operation of the advice transmitting unit 27 will be described here.

The advice sending unit 27 acquires advice to the care provider from the advice determining unit 23 .
The advice sending unit 27 gives advice to the learning model 47 and obtains, from the learning model 47, information indicating the timing of sending the advice.
If the advice is urgent, such as "Please cool Mr./Ms. X to bring down his/her fever," or "Check for chills," information indicating the timing of transmission is obtained from learning model 47, for example, information indicating that the advice will be sent immediately.
If the advice is not urgent, such as "Mr./Ms. XX may not be getting enough sleep," or "Please ask Mr./Ms. XX if he/she has any concerns," information indicating the timing of transmission is obtained from the learning model 47, for example, information indicating that the advice will be sent within one hour.
The advice sending unit 27 sends the advice to the chatbot engine unit 13 of the chatbot 1 at the determined timing.

In the above-described fourth embodiment, the advice transmitting device 2 is configured so that the advice transmitting unit 27 determines the timing for transmitting the advice based on the advice determined by the advice determining unit 23, and transmits the advice at the determined timing. Therefore, the advice transmitting device 2 can provide advice to the care provider at an appropriate timing when the situation of the care recipient changes.

In the advice sending device 2 shown in FIG. 1, the advice sending unit 27 gives advice to the learning model 47 and obtains information indicating the timing of sending the advice from the learning model 47. However, this is merely an example, and the advice sending unit 27 may determine the timing of sending the advice based on, for example, a rule base.

In addition, this disclosure allows for any combination of the embodiments, any modification of any component of each embodiment, or any omission of any component of each embodiment.

The present disclosure is suitable for an advice sending device and an advice sending method.

1 Chatbot, 2 Advice transmission device, 3 Sensing device, 11 Voice recognition unit, 12 Voice database, 13 Chatbot engine unit, 14 Display unit, 15 Voice output unit, 21 Text data acquisition unit, 22 Sensing data collection unit, 23 Advice determination unit, 24 Advice transmission unit, 25, 26 Advice determination unit, 27 Advice transmission unit, 31 Text data acquisition circuit, 32 Sensing data collection circuit, 33 Advice determination circuit, 34 Advice transmission circuit, 35, 36 Advice determination circuit, 37 Advice transmission circuit, 41, 42, 43, 44 Learning model, 45 Care record database, 46 Medical care information database, 47 Learning model, 51 Memory, 52 Processor, 61 Bed sensor, 62 Room sensor, 63 Home appliance sensor, 64 Human presence sensor, 65 Vital sensor, 66 Ingestion detection sensor, 67 Sensing data recording unit.

Claims (6)

a text data acquisition unit that acquires a notification request for advice to a care provider who provides care services to a certain care recipient among one or more care recipients, and outputs a command to collect sensing data related to the certain care recipient;
a sensing data collection unit that collects sensing data related to the certain care recipient from sensing data of a sensor that monitors the status of each care recipient according to a collection command output from the text data acquisition unit;
an advice determination unit that determines advice to a care provider who provides care services to the care recipient based on the sensing data collected by the sensing data collection unit;
an advice sending unit that sends the advice determined by the advice determining unit ,
The advice determination unit,
determining the necessity of the advice based on the sensing data collected by the sensing data collection unit, and if the necessity of the advice is present, determining the advice based on the sensing data;
The advice determination unit,
providing the sensing data collected by the sensing data collection unit to a first learning model realized by a neural network, and obtaining a determination result of the necessity of advice from the first learning model;
If the result of the determination of the necessity of advice indicates that there is a need for advice, the sensing data collected by the sensing data collection unit is provided to a second learning model realized by a neural network, and the advice corresponding to the sensing data is obtained from the second learning model;
the first learning model is a model that is learned to output a determination result of the necessity of advice in response to an input of sensing data collected by the sensing data collection unit from the advice determination unit,
The advice sending device is characterized in that the second learning model is a model that is learned to output advice to a care provider corresponding to sensing data input from the advice determination unit and collected by the sensing data collection unit . The advice determination unit,
The advice sending device of claim 1, characterized in that the need for the advice is determined based on the sensing data collected by the sensing data collection unit and the care records of the care recipient, and if there is a need for the advice, the advice is decided based on the sensing data and the care records. The advice determination unit,
The advice sending device of claim 1, characterized in that the need for the advice is determined based on the sensing data collected by the sensing data collection unit and the medication information of the care recipient, and if the need for the advice is determined, the advice is decided based on the sensing data and the medication information. The advice determination unit,
2. The advice transmitting device according to claim 1, further comprising: determining advice to the care provider based on a plurality of pieces of sensing data collected by the sensing data collecting unit. The advice sending unit is
2. The advice transmitting device according to claim 1, further comprising: determining a timing for transmitting the advice based on the advice determined by the advice determining section; and transmitting the advice at the determined timing. An advice transmitting method for an advice transmitting device including a text data acquiring unit, a sensing data collecting unit, an advice determining unit, and an advice transmitting unit, comprising:
The text data acquisition unit acquires a notification request for advice to a care provider who provides care services to a certain care recipient among one or more care recipients, and outputs a command to collect sensing data related to the certain care recipient;
the sensing data collection unit collects sensing data related to the certain care recipient from sensing data of sensors monitoring the status of each care recipient in accordance with the collection command output from the text data acquisition unit;
The advice determination unit determines advice to a care provider who provides a care service to the care recipient based on the sensing data collected by the sensing data collection unit;
the advice sending unit sends the advice determined by the advice determination unit ;
The advice determination unit,
determining the necessity of the advice based on the sensing data collected by the sensing data collection unit, and if the necessity of the advice is present, determining the advice based on the sensing data;
The advice determination unit,
providing the sensing data collected by the sensing data collection unit to a first learning model realized by a neural network, and obtaining a determination result of the necessity of advice from the first learning model;
If the result of the determination of the necessity of advice indicates that there is a need for advice, the sensing data collected by the sensing data collection unit is provided to a second learning model realized by a neural network, and the advice corresponding to the sensing data is obtained from the second learning model;
the first learning model is a model that is learned to output a determination result of the necessity of advice in response to an input of sensing data collected by the sensing data collection unit from the advice determination unit,
An advice sending method characterized in that the second learning model is a model trained to output advice to a care provider corresponding to sensing data input from the advice determination unit and collected by the sensing data collection unit .

Images