JP7587804B2 - Injury/disease estimation system, injury/disease estimation method, and injury/disease estimation program - Google Patents

Description

The present invention relates to an injury/illness estimation system, an injury/illness estimation method, and an injury/illness estimation program for estimating injuries and illnesses.

Traditionally, blood glucose level management has been used to maintain and manage a person's health. For example, Patent Document 1 discloses a blood glucose level prediction device that predicts a user's blood glucose level.

Patent No. 6659049

Incidentally, an increase in blood glucose level occurs not only as a result of food intake, but also when adrenaline secreted from the adrenal medulla acts on the liver to break down glycogen. It is generally known that blood glucose levels that are too high or too low have adverse effects on the human body. Although the technology of Patent Document 1 above can predict blood glucose levels, there is a problem in that it is not possible to predict what effects the predicted blood glucose levels will have on the human body. In addition, it is not possible to predict the effects on the human body of the secretion of adrenaline, which is associated with an increase in blood glucose levels.

Therefore, the present invention has been made in consideration of the above problems, and aims to provide an injury/illness estimation system, injury/illness estimation method, and injury/illness estimation program that can estimate injuries or illnesses that a user may suffer from by utilizing the user's blood glucose level.

An injury/illness estimation system according to one aspect of the present invention includes a memory unit that stores a learning model that learns the relationship between a user's blood glucose level during exercise and the user's injury or illness, a reception unit that receives input of blood glucose level information related to the target user's blood glucose level during exercise, an estimation unit that estimates the possibility that the target user will suffer an injury or illness using the blood glucose level information received by the reception unit and the learning model, and an output unit that outputs information related to the possibility estimated by the estimation unit.

In addition, in the above injury/illness estimation system, the learning model is a learning model that learns the relationship between the change in blood glucose level during exercise of the user and the user's injury/illness, and the reception unit may receive input of blood glucose level information regarding the change in blood glucose level during exercise of the target user.

In the above injury/illness estimation system, the learning model may further be a model that has learned the relationship between the user's redox potential during exercise and the user's injury or illness, and the reception unit may further receive input of redox potential information relating to the target user's redox potential during exercise, and the estimation unit may use the blood glucose level information and redox potential information received by the reception unit and the learning model to estimate the possibility that the target user will suffer an injury or illness.

An injury/illness estimation system according to one embodiment of the present invention includes a memory unit that stores a learning model that learns the relationship between the amount of adrenaline secreted by a user during exercise and the user's injury or illness, a reception unit that receives input of adrenaline information related to the amount of adrenaline secreted by a target user during exercise, an estimation unit that estimates the possibility that the target user will become injured or ill using the adrenaline information received by the reception unit and the learning model, and an output unit that outputs information related to the possibility estimated by the estimation unit.

In addition, in the above injury/illness estimation system, the amount of adrenaline secretion may indicate a value obtained by converting the blood glucose level measured while the user is exercising into adrenaline through a specified calculation.

In addition, in the above injury/illness estimation system, the learning model is a learning model that learns the relationship between the change in adrenaline secretion during exercise by the user and the user's injury or illness, and the reception unit may receive input of adrenaline information regarding the change in adrenaline secretion during exercise by the user.

In addition, in the above injury/illness estimation system, the memory unit stores a second learning model that has learned the relationship between blood glucose level and adrenaline, and includes an acquisition unit that acquires blood glucose level information that measures the user's blood glucose level, and a second estimation unit that estimates the user's adrenaline secretion amount based on the blood glucose level information acquired by the acquisition unit and the learning model, and the reception unit may be configured to receive the adrenaline secretion amount estimated by the second estimation unit.

In addition, in the above injury/illness estimation system, the learning model may be a model that has learned the relationship between the user's redox potential during exercise and the user's injury or illness, and the reception unit may further receive input of redox potential information related to the target user's redox potential during exercise, and the estimation unit may use the adrenaline information and redox potential information received by the reception unit and the learning model to estimate the possibility that the target user will suffer an injury or illness.

In addition, an injury/illness estimation method according to one aspect of the present invention executes a reception step in which a computer receives input of blood glucose level information relating to the blood glucose level of a target user during exercise, an estimation step in which the computer estimates the possibility of the target user suffering from an injury or illness using the blood glucose level information received by the reception unit and a learning model that has learned the relationship between the user's blood glucose level during exercise and the user's injury or illness, and an output step in which the computer outputs information relating to the possibility estimated in the estimation step.

In addition, an injury/illness estimation program according to one embodiment of the present invention provides a computer with a reception function for receiving input of blood glucose level information relating to the blood glucose level of the target user during exercise, an estimation function for estimating the possibility that the target user will suffer an injury or illness using the blood glucose level information received by the reception unit and a learning model that has learned the relationship between the user's blood glucose level during exercise and the user's injury or illness, and an output function for outputting information relating to the possibility estimated by the estimation function.

In addition, an injury/illness estimation method according to one aspect of the present invention executes a reception step in which a computer receives input of adrenaline information related to the amount of adrenaline secreted by the target user during exercise, an estimation step in which the computer estimates the possibility that the target user will become injured or ill using the adrenaline information received in the reception step and a learning model that has learned the relationship between the amount of adrenaline secreted by the user during exercise and the user's injury or illness, and an output step in which the computer outputs information related to the possibility estimated in the estimation step.

In addition, an injury/illness estimation program according to one embodiment of the present invention provides a computer with a reception function for receiving input of adrenaline information related to the amount of adrenaline secreted by the target user during exercise, an estimation function for estimating the possibility that the target user will become injured or ill using the adrenaline information received by the reception function and a learning model that has learned the relationship between the amount of adrenaline secreted by the user during exercise and the user's injury or illness, and an output function for outputting information related to the possibility estimated by the estimation function.

An adrenaline estimation system according to one aspect of the present invention includes a memory unit that stores a learning model that learns the relationship between blood glucose level and adrenaline, an acquisition unit that acquires blood glucose level information obtained by measuring the blood glucose level of a user, an estimation unit that estimates the user's adrenaline secretion amount based on the blood glucose level information acquired by the acquisition unit and the learning model, and an output unit that outputs the adrenaline secretion amount estimated by the estimation unit.

In addition, an adrenaline estimation method according to one aspect of the present invention executes an acquisition step in which a computer acquires blood glucose level information that measures a user's blood glucose level, an estimation step in which adrenaline secretion amount of the user is estimated based on the blood glucose level information acquired in the acquisition step and a learning model that has learned the relationship between blood glucose level and adrenaline, and an output step in which adrenaline secretion amount estimated in the estimation step is output.

In addition, an adrenaline estimation program according to one embodiment of the present invention causes a computer to realize an acquisition function for acquiring blood glucose level information that measures a user's blood glucose level, an estimation function for estimating the user's adrenaline secretion amount based on the blood glucose level information acquired by the acquisition function and a learning model that has learned the relationship between blood glucose level and adrenaline, and an output function for outputting the adrenaline secretion amount estimated by the estimation function.

An injury/illness estimation system according to one aspect of the present invention uses a learning model that learns the relationship between a user's blood glucose level during exercise and the user's injury or illness, and by obtaining the blood glucose levels of other users during exercise, can estimate the possibility that the user will become injured or ill. As a result, for example, information for improving the user's lifestyle and exercise can be obtained from the estimated content.

FIG. 1 is a system diagram showing an overview of an injury/illness estimation system. FIG. 2 is a block diagram showing an example of the configuration of an injury/illness estimation server. FIG. 2 is a conceptual data diagram showing an example of training data for generating an injury/illness estimation learning model. FIG. 2 is a block diagram showing an example of the configuration of an adrenaline estimation server. FIG. 11 is a conceptual data diagram showing an example of training data for generating an adrenaline estimation learning model. FIG. 2 is a block diagram showing a configuration example of a sensor. FIG. 4 is a diagram showing an example of a specific configuration of a detection unit of a sensor. FIG. 2 is a sequence diagram showing an example of communication between devices related to the injury/illness inference system. 11 is a flowchart showing an example of the operation of the sensor. 13 is a flowchart showing an example of the operation of the adrenaline estimation server. 13 is a flowchart showing an example of the operation of the injury/disease estimation server. FIG. 13 is a diagram illustrating an example of a display on a user terminal. FIG. 13 is a system diagram showing another example configuration of the injury/illness inference system. FIG. 2 is a conceptual data diagram showing an example of training data for generating an injury/illness estimation learning model. 13 is a flowchart showing an example of the operation of the injury/disease estimation server. FIG. 13 is a diagram showing an example of another specific configuration of the detection unit of the sensor. FIG. 13 is a diagram showing an example of another specific configuration of the detection unit of the sensor. FIG. 2 is a conceptual data diagram showing an example of training data for generating an injury/illness estimation learning model. FIG. 2 is a sequence diagram showing an example of communication between devices related to the injury/illness inference system.

First Embodiment
Fig. 1 is a system diagram showing an overview of an injury/illness estimation system according to a first embodiment. As shown in Fig. 1, the injury/illness estimation system includes an injury/illness estimation server 100, an adrenaline estimation server 200, and may include a sensor 300 and a user terminal 500. The injury/illness estimation server 100, the adrenaline estimation server 200, the sensor 300, and the user terminal 500 may be communicably connected to each other via a network 400. Although one of each device is shown in Fig. 1, the number of each terminal and server is not limited to one, and there may be multiple terminals and servers.

The injury/illness estimation server 100 is a server device (information processing device) that estimates the possibility that the user 30 will become injured or ill from the amount of adrenaline secreted by the user 30 while exercising. The adrenaline estimation server 200 is a server device (information processing device) that estimates the amount of adrenaline secreted by the user 30 from the blood glucose level (amount of glucose secreted) while exercising. The sensor 300 is a sensor that detects and outputs the amount of glucose secreted from the bodily fluids of the user 30. The user terminal 500 is an information processing terminal held by the user 30, and may be realized by a so-called smartphone, tablet terminal, mobile phone, PC, etc.

The injury/illness estimation system shown in FIG. 1 is an estimation system that estimates the possibility that a user 30 has an injury or illness or will suffer an injury or illness based on the amount of adrenaline secreted by the user 30 while exercising. As shown in FIG. 1, the user 30 measures his/her blood glucose level while exercising using a sensor 300. The sensor 300 transmits the measured sensing data 31 to the adrenaline estimation server 200. The adrenaline estimation server 200 estimates the amount of adrenaline secreted by the user 30 based on the received sensing data 31 indicating the blood glucose level of the user 30. The adrenaline estimation server 200 transmits adrenaline information 21 indicating the estimated amount of adrenaline secreted to the injury/illness estimation server 100. The injury/illness estimation server 100 estimates the possibility that the user 30 will suffer (or has suffered) an injury or illness based on the received adrenaline information 21. Then, the injury/illness estimation server 100 transmits injury/illness estimation information 11 indicating the likelihood of the user 30 suffering from the estimated injury or illness to the user terminal 500 of the user 30. The user terminal 500 displays the received injury/illness estimation information 11. This allows the user 30 to recognize the likelihood that the user 30 will suffer from an injury or illness.

This will be explained in detail below. The network 400 is a network for connecting various devices to each other, for example, a wireless network or a wired network. Specifically, the network is a wireless LAN (WLAN), a wide area network (WAN), ISDNs (integrated service digital networks), wireless LANs, LTE (long term evolution), LTE-Advanced, 4th generation (4G), 5th generation (5G), CDMA (code division multiple access), WCDMA (registered trademark), Ethernet (registered trademark), etc. The network is not limited to these examples, and may be, for example, a Public Switched Telephone Network (PSTN), Bluetooth (registered trademark), Bluetooth Low Energy, an optical line, an Asymmetric Digital Subscriber Line (ADSL) line, a satellite communication network, or any other network. When the network is provided in a user's residence, it may be called a home network. The network may be, for example, a Narrow Band IoT (NB-IoT) or an enhanced Machine Type Communication (eMTC). Note that NB-IoT and eMTC are wireless communication methods for IoT, and are networks that enable long-distance communication at low cost and low power consumption. The network may also be a combination of these. The network may also include multiple different networks that combine these examples. For example, the network may include a wireless network using LTE and a wired network such as an intranet, which is a closed network.

<Configuration example of injury/illness estimation server 100>
2 is a block diagram showing an example of the configuration of the injury/illness estimation server 100. As described above, the injury/illness estimation server 100 is a server device (information processing device) that estimates the possibility that a user will become injured or ill, based on the amount of adrenaline secreted during exercise by the user.

As shown in FIG. 2, the injury/illness estimation server 100 includes a communication unit 110, an input unit 120, a control unit 130, a memory unit 140, and an output unit 150. The communication unit 110, the input unit 120, the control unit 130, the memory unit 140, and the output unit 150 may be configured to be able to communicate with each other via a bus 160.

The communication unit 110 is a communication interface having a function for communicating with other devices. The communication unit 110 may communicate with other devices using any communication protocol, and may communicate either wired or wirelessly, as long as it is capable of communicating with other devices. The communication unit 110 communicates with the adrenaline estimation server 200 and the user terminal 500 according to instructions from the control unit 130. For example, the communication unit 110 receives adrenaline information 21 transmitted from the adrenaline estimation server 200 via the network 400. The communication unit 110 also transmits injury/illness estimation information 11 indicating the possibility that the user 30 will suffer an injury or illness, as estimated by the injury/illness estimation server 100, to the user terminal 500 via the network 400.

The input unit 120 is an input interface that has the function of accepting input from an operator of the injury/illness estimation server 100, etc., and transmitting the input to the control unit 130. The input unit 120 may be realized by soft keys such as a touch panel, or may be realized by hard keys. Alternatively, the input unit 120 may be a microphone for accepting voice input. The input unit 120 may also be a port for accepting input from a storage medium such as a flash memory. The input unit 120 accepts direct input of adrenaline information (e.g., the numerical value of the amount of adrenaline secreted per hour) and transmits it to the control unit 130.

The memory unit 140 has a function of storing various programs and data required for the operation of the injury/illness estimation server 100. The memory unit 140 can be realized, for example, by a hard disk drive (HDD), a solid state drive (SSD), flash memory, etc. The memory unit 140 stores an injury/illness estimation learning model 141. The injury/illness estimation learning model 141 may be a model generated by the injury/illness estimation server 100, or may store one generated by another device.

The injury/illness estimation learning model 141 is a learning model that has learned the relationship between the change in adrenaline secretion during exercise and the injury or illness suffered by the user, and is a model for obtaining the adrenaline secretion amount during exercise of the user and estimating the possibility that the user will contract an injury or illness (or the possibility that the user has contracted an injury or illness). The injury/illness estimation learning model 141 may output the names of injuries or illnesses that the user may have, or may output the names of injuries or illnesses that the user is more than a certain possibility of having, or may output a numerical value (e.g., a percentage) or an evaluation (e.g., evaluation of large, medium, small, etc.) indicating the possibility that the user will have each injury or illness.

Figure 3 is a conceptual data diagram showing an example of teacher data for generating the injury/illness estimation learning model 141. As shown in Figure 3, teacher data 1410 is information in which subject ID 1411, gender 1412, weight 1413, adrenaline data 1414, and disease information 1415 are associated with each other. Teacher data 1410 is a collection of information in which the amount of adrenaline secreted during exercise by the user is associated with disease information indicating the injury or illness suffered by the user.

The subject ID 1411 is information that allows the injury/illness estimation server 100 to identify each piece of teacher data in the injury/illness estimation system, and is information that identifies the user (subject).
The gender 1412 is information indicating the gender of the user identified by the corresponding subject ID 1411 .
The weight 1413 is information indicating the weight of the user identified by the corresponding subject ID 1411 .

Adrenaline data 1414 is information indicating the amount of adrenaline secreted during exercise by the user indicated by the corresponding subject ID 1411. This amount of adrenaline secreted is information estimated from the user's blood glucose level by the adrenaline estimation server 200. For convenience, FIG. 3 shows identification information indicating the information that holds the adrenaline data.

The disease information 1415 is information indicating an injury or illness that the user indicated by the corresponding subject ID 1411 has suffered from or is suffering from. The disease information 1415 may include information on multiple injuries or illnesses. In addition, when only one injury or illness is registered as the disease information 1415, other injuries or illnesses of the same user may be registered as separate data. The disease information 1415 may further include information on the date on which the disease or illness was contracted. If the injury or illness currently suffered by the corresponding user is associated as the disease information 1415, the injury or illness estimation server 100 can estimate the possibility that the user is currently suffering from an injury or illness. In addition, if information on an injury or illness suffered several years (e.g., 10 years) after the date on which the blood glucose level information that is the basis of the corresponding adrenaline data 1414 was measured is used as the disease information 1415, the injury or illness estimation server 100 can estimate an injury or illness that the user may suffer in the future.

The teacher data 1410 may include at least adrenaline data 1414 and disease information 1415, but may not include other information. Conversely, information about the subject corresponding to the adrenaline data 1414 may include information other than gender and weight, and may be associated with information such as the user's height, build, and BMI (Body Mass Index).

Returning to FIG. 2, the output unit 150 has a function of outputting specified information in accordance with instructions from the control unit 130. The output by the output unit 150 may be either an image signal or an audio signal. In the case of outputting an image signal, the output may be to a monitor connected to (or provided by) the injury/illness estimation server 100. In the case of outputting an audio signal, the output may be to a speaker connected to (or provided by) the injury/illness estimation server 100. The output unit 150 outputs, for example, information indicating the possibility that the user may become injured or ill.

The control unit 130 is a processor having the function of controlling each part of the injury/illness estimation server 100. The control unit 130 may be realized by a single core or by a multi-core. The control unit 130 realizes the functions of the injury/illness estimation server 100 by executing various programs stored in the memory unit 140.

The control unit 130 includes, as functions realized by the control unit 130, a reception unit 131 and an estimation unit 132. The control unit 130 may also include a learning unit 133.

The reception unit 131 receives adrenaline information of a user who is to be estimated as to whether or not the user is likely to become injured or ill, and transmits the information to the estimation unit 132. The reception unit 131 receives the user's adrenaline information from the communication unit 110 or from the input unit 120. The reception unit 131 transmits the received adrenaline information to the estimation unit 132.

The estimation unit 132 receives the adrenaline information transmitted from the reception unit 131 as input, and uses the injury/illness estimation learning model 141 stored in the memory unit 140 to estimate whether the user corresponding to the input adrenaline information is likely to suffer from some injury or illness. The estimation unit 132 may estimate the name of an injury or illness that the user may suffer from, and the estimated injury or illness name may be one or more. If there is no possibility of suffering from an injury or illness, the estimation unit 132 may estimate that there is no possibility of suffering from an injury or illness. Furthermore, the estimation unit 132 may estimate the injury or illness that the user is most likely to suffer from, or may estimate an injury or illness that has a predetermined or higher probability of suffering (e.g., 30% or higher).

The learning unit 133 learns the relationship between the amount of adrenaline secreted during exercise by the user and the injury or illness suffered by the user. Specifically, the learning unit 133 receives input of teacher data as shown in FIG. 3, performs learning using the disease information 1415 as a so-called label, and generates an injury or illness estimation learning model 141. An existing learning algorithm may be used by the learning unit 133. The learning unit 133 stores the generated injury or illness estimation learning model 141 in the storage unit 140. The learning unit 133 may perform re-learning using the injury or illness estimation learning model 141 stored in the storage unit 140 and newly input teacher data. In addition, the learning unit 133 may perform re-learning by receiving information regarding the truth or falsity of the estimation based on the injury or illness estimation information provided to the user as feedback information.
The above is a description of an example of the configuration of the injury/disease estimation server 100.

<Adrenaline Estimation Server 200>
4 is a block diagram showing an example of the configuration of the adrenaline estimation server 200. As described above, the adrenaline estimation server 200 is a server device (information processing device) that estimates the amount of adrenaline secreted based on the blood glucose level information of the user. It is generally known that when adrenaline is secreted in humans, blood glucose levels rise. In other words, it can be said that there is a close relationship between blood glucose levels and adrenaline. Therefore, conversely, it is also possible to estimate the amount of adrenaline secreted from blood glucose levels.

As shown in FIG. 4, the adrenaline estimation server 200 includes a communication unit 210, an input unit 220, a control unit 230, a storage unit 240, and an output unit 250. The communication unit 210, the input unit 220, the control unit 230, the storage unit 240, and the output unit 250 may be configured to be able to communicate with each other via a bus 260.

The communication unit 210 is a communication interface having a function for communicating with other devices. The communication unit 210 may communicate with other devices using any communication protocol as long as it is capable of communicating with other devices, and the communication may be wired or wireless. The communication unit 210 communicates with the injury/illness estimation server 100 according to instructions from the control unit 230. The communication unit 210 may also communicate with the sensor 300.

The input unit 220 is an input interface that has the function of accepting input from an operator of the adrenaline estimation server 200 and transmitting it to the control unit 230. The input unit 220 may be realized by soft keys such as a touch panel, or may be realized by hard keys. Alternatively, the input unit 220 may be a microphone for accepting voice input. The input unit 220 transmits the input contents entered by the operator to the control unit 230. The input unit 220 accepts, for example, direct input of blood glucose level information (e.g., blood glucose value for each hour) and transmits it to the control unit 230.

The memory unit 240 has a function of storing various programs and data required for the operation of the adrenaline estimation server 200. The memory unit 240 can be realized, for example, by a hard disk drive (HDD), a solid state drive (SSD), flash memory, etc. The memory unit 240 stores an adrenaline estimation learning model 241. The adrenaline estimation learning model 241 may be generated by the adrenaline estimation server 200, or may store one generated by another device.

The adrenaline estimation learning model 241 is a learning model that has learned the relationship between the user's blood glucose level during exercise and the amount of adrenaline secreted, and is a model for obtaining the user's blood glucose level during exercise and estimating the amount of adrenaline secreted that the user is believed to be secreting. The adrenaline estimation learning model 241 estimates the qualitative amount of adrenaline secreted.

Figure 5 is a conceptual data diagram showing an example of teacher data for generating an adrenaline estimation learning model 241. As shown in Figure 5, teacher data 2410 is information in which subject ID 2411, gender 2412, weight 2413, blood glucose data 2414, and adrenaline data 2415 are associated with each other. Teacher data 2410 is a collection of information in which the amount of adrenaline secreted during exercise by the user is associated with disease information indicating an injury or illness suffered by the user.

The subject ID 2411 is information that the adrenaline estimation server 200 uses in the injury/illness estimation system to identify each piece of teacher data, and is information that identifies the user (subject).

The gender 2412 is information indicating the gender of the user identified by the corresponding subject ID 2411 .
The weight 2413 is information indicating the weight of the user identified by the corresponding subject ID 2411 .

The blood glucose level data 2414 is information indicating the blood glucose level during exercise of the user indicated by the corresponding subject ID 2411. This blood glucose level may be measured by the sensor 300 or may be measured by another sensor. For convenience, FIG. 5 shows identification information indicating information that holds the blood glucose level value.

The adrenaline data 2415 is information indicating the amount of adrenaline secreted during exercise by the user indicated by the corresponding subject ID 2411. This amount of adrenaline secreted is information estimated from the user's blood glucose level by the adrenaline estimation server 200. For convenience, FIG. 5 shows identification information indicating the information that holds the adrenaline data.

The teacher data 2410 may include at least blood glucose data 2414 and adrenaline data 2415, and may not include other information. Conversely, the information on the subject corresponding to the blood glucose data 2414 may include information other than gender and weight, and may correspond to information such as the user's height, build, and BMI (Body Mass Index). In this case, the adrenaline data 2415 used as teacher data may be data of a value corresponding to the amount of blood glucose at a time when a state or emotional expression occurs in which it can be estimated that adrenaline is being secreted based on the change in the time series indicated by the corresponding blood glucose data 2414, and may be information set so that the amount of adrenaline secreted is low even if the blood glucose level is high at a time when it can be estimated that adrenaline is not being secreted.

Returning to FIG. 4, the output unit 250 has a function of outputting specified information in accordance with instructions from the control unit 230. The output by the output unit 250 may be either an image signal or an audio signal. In the case of outputting an image signal, the output may be to a monitor connected to the adrenaline estimation server 200 (or provided in the adrenaline estimation server 200). In the case of outputting an audio signal, the output may be to a speaker connected to the adrenaline estimation server 200 (or provided in the adrenaline estimation server 200). The output unit 250 outputs, for example, adrenaline information indicating the amount of adrenaline secretion estimated from the user's blood glucose level.

The control unit 230 is a processor that has the function of controlling each part of the adrenaline estimation server 200. The control unit 230 may be realized by a single core or by a multi-core. The control unit 230 realizes the functions of the adrenaline estimation server 200 by executing various programs stored in the memory unit 240.

The control unit 230 includes, as functions realized by the control unit 230, a reception unit 231 and an estimation unit 232. The control unit 230 may also include a learning unit 233.

The reception unit 231 receives blood glucose level information of a user who is to be estimated as to whether or not the user is likely to become ill, and transmits the information to the estimation unit 232. The reception unit 231 receives the blood glucose level information of the user by being transmitted from the communication unit 210 or by being transmitted from the input unit 220. The reception unit 231 transmits the received adrenaline information to the estimation unit 232.

The estimation unit 232 receives the blood glucose level information transmitted from the reception unit 231 as input, and uses the adrenaline estimation learning model 241 stored in the memory unit 240 to estimate the user's adrenaline secretion amount corresponding to the input blood glucose level information.

The learning unit 233 learns the relationship between the amount of adrenaline secreted during exercise by the user and the injury or illness suffered by the user. Specifically, the learning unit 233 receives input of teacher data as shown in FIG. 3, performs learning using the adrenaline data 2415 as a so-called label, and generates the adrenaline estimation learning model 241. The learning unit 233 may use an existing learning algorithm. The learning unit 233 stores the generated adrenaline estimation learning model 241 in the storage unit 240. The learning unit 233 may perform re-learning using the adrenaline estimation learning model 241 stored in the storage unit 240 and newly input teacher data.
The above is a description of an example of the configuration of the adrenaline estimation server 200.

<Configuration example of sensor 300>
6 is a block diagram showing an example of the configuration of the sensor 300. As shown in FIG. 6, the sensor 300 includes a communication unit 310, a detection unit 320, a storage unit 340, and an output unit 350.

The communication unit 310 is a communication interface having a function for communicating with other devices. The communication unit 310 may communicate with other devices using any communication protocol as long as the communication unit 310 can communicate with other devices, and may communicate either wired or wirelessly. The communication unit 310 communicates with the adrenaline estimation server 200. The communication unit 310 may also communicate with the user terminal 500.
The detection unit 320 detects the blood glucose level of the user using the bodily fluids (blood) of the user wearing the sensor 300.
FIG. 7 is a diagram showing an example of a specific configuration of the detection unit 320. As shown in FIG.

Figure 7 is a schematic diagram of the main components of the detection unit 320 of the first application form. The detection unit 320 has a detection element 321 of a predetermined container shape, which is connected by wiring to the measurement element 700. A bodily fluid sampling section 330 is formed in the detection element 321. The bodily fluid sampling section 330 has a known needle shape, and biological fluids 322 such as blood, sweat, and saliva are sampled by surface tension and guided into the detection element 321.

The detection element 321 contains a receptor for identifying at least glucose as the detection target substance and detecting the glucose concentration in the body fluid collected from the body fluid collection section 330. The receptor is formed of a self-assembled monolayer (SAM) containing an identifying substance and an inhibitor. A SAM is an organic thin film in which organic molecules spontaneously assemble together at the interface between a solid and a liquid or between a solid and a gas to spontaneously form a monolayer.

The identification substance has a function of binding to glucose contained in a sample. Phenylboronic acid can be used as the identification substance, and a phenylboronic acid derivative is particularly preferred. This phenylboronic acid derivative has a halogen group such as F (fluorine) or Cl (chlorine) or NO ₂ (nitro group) in the aromatic ring portion of the phenylboronic acid. The use of a phenylboronic acid derivative having a nitro group in the aromatic ring portion is more preferable than the use of a halogen group, since it has a significant reactivity with glucose.

The inhibitor prevents proteins such as albumin, which are not the detection target substances, from binding to the phenylboronic acid derivative or from adhering to the surface of the redox potential measurement electrode 326. In this embodiment, the inhibitor is formed of a polymer compound. The polymer compound may be oligoethylene glycol, which has a molecular chain longer than the identification substance, or, for example, polyethylene glycol.

An oxidation-reduction potential measuring electrode 326 and a reference electrode 329 are attached in the detection element 321. A glucose capture molecule 327 (phenylboronic acid derivative) for capturing glucose 323 as the detection target substance is fixed to the oxidation-reduction potential measuring electrode 326. Since the biological fluid 322 is blood (or may be sweat, saliva, etc.), in addition to glucose 323 as the detection target substance, oxidation-reduction potential fluctuation factors 324 and impurities 325 are also present. The oxidation-reduction potential fluctuation factors 324 and impurities 325 are proteins, garbage, etc. In addition, in the detection unit 320, substances other than glucose may also be detected through the oxidation-reduction potential measuring electrode 326. For this, an antibody or aptamer that captures the detection target substance is used as the capture molecule 327.

One end of the glucose capture molecule 327 is electrically connected to the electrode 328. The measurement element 700 is realized by an FET, for example. Glucose in the sample binds to the glucose capture molecule 327 (phenylboronic acid derivative) as an identification substance. As a result, the glucose capture molecule 327 generates a negative charge. This negative charge is applied to the electrode 328. When the measurement element 700 is an FET, it measures the change in the gate voltage that changes with the charge of the electrode 328. In this way, the detection unit 320 detects the glucose concentration contained in the sample. Alternatively, the measurement element 700 may be configured to detect the glucose concentration based on the potential difference between the reference electrode 329 and the electrode 328. Note that the configuration of the detection unit 320 shown in FIG. 7 is merely an example, and other configurations may be used as long as the glucose concentration can be detected.

Returning to FIG. 6, the memory unit 340 stores sensing data 341, which is blood glucose level information detected by the detection unit 320.

The output unit 350 outputs the sensing data 341 stored in the memory unit 340 or the blood glucose level information detected by the detection unit. The output unit 350 can be realized, for example, as an output terminal such as a USB terminal.

Note that the sensor 300 need only be able to output the detected blood glucose level, and need not include the communication unit 310. For example, the output unit 350 may be implemented as a USB port, and the detected blood glucose level information may be output by connecting to a user terminal 500 or the like. In such a case, the blood glucose level information transmitted to the adrenaline estimation server 200 may be transmitted from the user terminal 500. Also, if the blood glucose level measured by the detection unit 320 is output directly from the communication unit 310 or the output unit 350, the sensor 300 may not include the memory unit 340.
The above is a description of the configuration example of the sensor 300.

<Operation>
FIG. 8 is a sequence diagram showing a flow of injury/illness estimation by the injury/illness estimation system, and an example of exchange between devices related to the injury/illness estimation system.

As shown in FIG. 8, the sensor 300 is attached to the user 30 and detects the user's blood glucose level (glucose concentration in the blood) from the user's blood (step S801). The sensor 300 transmits the detected blood glucose level as blood glucose level information to the adrenaline estimation server 200 (step S802).

The adrenaline estimation server 200 estimates the amount of adrenaline secreted by the user 30 from the received blood glucose level information (step S803). The adrenaline estimation server 200 transmits adrenaline information indicating the estimated amount of adrenaline secreted by the user 30 to the injury/illness estimation server 100 (step S804).

When the injury/illness estimation server 100 receives the adrenaline information, it estimates the possibility that the user 30 will suffer an injury or illness based on the received adrenaline information (step S805). The injury/illness estimation server 100 transmits injury/illness estimation information indicating the estimated possibility of suffering an injury or illness to the user terminal 500 (step S806).

The user terminal 500 of the user 30 displays the received injury/illness estimation information (step S807). This allows the user 30 to recognize the possibility that the user 30 will become injured or ill, or the possibility that the user 30 has become injured or ill.
FIG. 9 is a flowchart showing an example of the operation of the sensor 300 for realizing the exchange shown in FIG.

As shown in FIG. 9, the detection unit 320 of the sensor 300 measures the glucose level of the user 30 (subject) (step S901). The detection unit 320 stores the measured glucose level in the storage unit 340.

The communication unit 310 transmits the sensing data stored in the storage unit 340 to the adrenaline estimation server (step S902). Note that the sensing data includes identification information (sensor ID) indicating the sensor 300 or identification information (subject ID) indicating the user 30 using the sensor 300.
The sensor 300 continues to execute the process shown in FIG. 9 until it is removed from the user 30 or the activation switch of the sensor 300 is turned off.
FIG. 10 is a flowchart showing an example of the operation of the adrenaline estimation server 200 for realizing the exchange shown in FIG.

As shown in FIG. 10, the communication unit 210 of the adrenaline estimation server 200 receives blood glucose level information indicating the blood glucose level of the user 30 transmitted from the sensor 300 (step S1001). The communication unit 210 transmits the received blood glucose level information to the control unit 230. The reception unit 231 of the control unit 230 accepts the blood glucose level information transmitted from the communication unit 210 as information on the subject whose possibility of suffering an injury or illness is to be estimated, and transmits it to the estimation unit 232.

When the blood glucose level information is transmitted, the estimation unit 232 uses the transmitted blood glucose level information as input and estimates the adrenaline secretion amount of the user 30 using the adrenaline estimation learning model 241 stored in the memory unit 240 (step S1002).

The estimation unit 232 then transmits adrenaline information indicating the estimated amount of adrenaline secretion to the injury/illness estimation server 100 via the communication unit 210 (step S1003), and ends the process. The adrenaline information transmitted from the adrenaline estimation server 200 to the injury/illness estimation server 100 includes identification information for identifying which user the information belongs to. The adrenaline estimation server 200 executes the process shown in FIG. 10 each time it receives blood glucose level information.
FIG. 11 is a flowchart showing an example of the operation of the injury/disease estimation server 100 for realizing the exchange shown in FIG.

As shown in FIG. 11, the communication unit 110 of the injury/illness estimation server 100 receives adrenaline information indicating the amount of adrenaline secretion estimated based on the blood glucose level of the user 30, transmitted from the adrenaline estimation server 200 (step S1101). The communication unit 110 transmits the received adrenaline information to the control unit 130. The reception unit 131 of the control unit 130 accepts the adrenaline information transmitted from the communication unit 110 as information on the subject whose possibility of suffering an injury or illness is estimated, and transmits it to the estimation unit 132.

When the adrenaline information is transmitted, the estimation unit 132 uses the transmitted adrenaline information as input and, using the injury/illness estimation learning model 141 stored in the memory unit 140, estimates the possibility that the user 30 will become injured or ill, or the possibility that the user 30 has an injury or illness (step S1102).

The estimation unit 132 then transmits injury/illness estimation information indicating the estimated possibility that the user 30 will become injured or ill to the user terminal 500 via the communication unit 110 (step S1103), and ends the process. The injury/illness estimation information transmitted from the injury/illness estimation server 100 to the user terminal 500 may include identification information for identifying which user the information belongs to. The injury/illness estimation server 100 executes the process shown in FIG. 10 every time it receives adrenaline information.
The above is the method used by the injury/illness inference system to estimate the possibility of a user becoming injured or ill based on the amount of adrenaline secreted during exercise.

Figure 12 is a diagram showing an example of the display of estimated injury/illness information for user 30 displayed on a user terminal 500 or the like. Figure 12 shows an example of the display when the user terminal 500 is assumed to be a smartphone.

As shown in FIG. 12, the user terminal 500 displays, as the injury/illness estimation information, injury/illness information 1212 indicating the possibility that the user 30, who is the holder of the user terminal 500, may suffer. FIG. 12 shows an example in which the injury/illness information 1212 displays the name of the injury or illness and the probability of suffering from that injury or illness, and shows an example in which there is an 80% possibility of becoming obese and a 40% possibility of developing diabetes. The user terminal 500 may also display, as the injury/illness estimation information, adrenaline information 1211 indicating the amount of adrenaline secreted by the user, which is the basis of the injury/illness information 1212. FIG. 12 shows an example in which the adrenaline information 1211 shows the progress of the amount of adrenaline secreted over time in a graph.

<Summary of First Embodiment>
As shown in the first embodiment, the injury/illness estimation server 100 can estimate the possibility that the user 30 will become injured or ill, based on the amount of adrenaline secreted by the user 30 during exercise. It is estimated that a large amount of adrenaline is secreted when the user 30 is exercising, particularly in a tense situation, and the blood glucose level increases accordingly. In addition, since adrenaline is involved in increasing the user's blood glucose level, it may affect the user's health condition, and it is possible to estimate the possibility that other users will become injured or ill, based on the amount of adrenaline secreted by other users during exercise and the history of injuries and illnesses suffered by the user. Therefore, a user who is estimated to have a possibility of suffering some injury or illness can take measures to prevent the injury or illness from occurring.

Second Embodiment
In the first embodiment, the possibility that the user 30 will become ill is estimated based on the amount of adrenaline secreted during exercise by the user, but this is not limited to the above. The injury/illness estimation server 100 may directly estimate an injury or illness from the blood glucose level (amount of glucose secreted) of the user 30, instead of the amount of adrenaline secreted by the user 30.

Figure 13 is a system diagram showing an example configuration of an injury/illness estimation system according to the second embodiment. As shown in Figure 13, the injury/illness estimation system includes an injury/illness estimation server 100, and may also include a sensor 300 and a user terminal 500. As shown in Figure 13, unlike the first embodiment, the injury/illness estimation system does not include an adrenaline estimation server 200.

In the injury/illness inference system shown in FIG. 13 , the sensor 300 transmits measured sensing data (blood glucose level information) directly to the injury/illness inference server 100. The injury/illness inference server 100 then infers the possibility that the user 30 will develop an injury or illness from the received blood glucose level information. The injury/illness inference server 100 then transmits the injury/illness inference information estimated based on the blood glucose level to the user terminal 500 of the user 30. In this way, in the second embodiment, the injury/illness inference system estimates the possibility that the user 30 will develop an injury or illness from the blood glucose level of the user 30 during exercise, without converting it into the amount of adrenaline secretion. A specific description will be given below, but in the second embodiment, only the differences from the first embodiment will be described.
The injury/disease estimation server 100 has the same configuration as that shown in Fig. 2. However, there are differences in the functions realized by each functional unit.
First, the receiving unit 131 receives blood glucose level information of the user 30 during exercise instead of adrenaline information of the user 30 during exercise, and transmits this to the estimation unit 132.

The estimation unit 132 also estimates the possibility that the user 30 will become ill, not from adrenaline information, but from blood glucose level information and an injury/illness estimation learning model. Here, the injury/illness estimation learning model according to the second embodiment will be described.

The injury/illness estimation learning model according to the second embodiment is a model that learns the relationship between blood glucose level information of a user during exercise and an injury or illness suffered by the user, and is a model that accepts input of blood glucose level information of a user 30 during exercise and is used to estimate the possibility (or the possibility that the user 30 has) that the user 30 will suffer an injury or illness.

14 is a data conceptual diagram showing an example of the data configuration of teacher data 3410 used to generate an injury/illness estimation learning model according to the second embodiment. The data configuration example of teacher data 3410 shown in FIG. 14 is roughly the same as teacher data 1410 shown in FIG. 3, but differs in that while teacher data 1410 shown in the first embodiment holds adrenaline data 1414, this second embodiment holds blood glucose data 2414 instead of adrenaline data 1414. Therefore, a model generated when learning using teacher data 3410 shown in FIG. 14 is a model that learns the relationship between blood glucose and the possibility of suffering an injury or illness. The teacher data 3410 only needs to correspond at least blood glucose data 2414 and disease information 1415, and other information may be deleted or other information related to the corresponding user may be added as necessary.

Figure 15 is a flowchart showing an example of the operation of the injury/illness estimation server 100 according to the second embodiment.

As shown in FIG. 15, the communication unit 110 of the injury/illness estimation server 100 receives blood glucose level information of the user 30 (subject) from the sensor 300 (step S1501). The communication unit 110 transmits the received blood glucose level information to the control unit 130.

The reception unit 131 of the control unit 130 receives the transmitted blood glucose level information as information for estimating the possibility that the user 30 will suffer an injury or illness. The reception unit 131 transmits the received blood glucose level information to the control unit 130. The estimation unit 132 of the control unit 130 uses the received blood glucose level information as input and estimates the possibility that the user 30 will suffer an injury or illness using the injury/illness estimation learning model 141 (step S1502).

The estimation unit 132 generates injury/illness estimation information indicating the possibility of suffering the estimated injury/illness, transmits the generated injury/illness estimation information to the user terminal 500 via the communication unit 110 (step S1503), and ends the processing.

In this way, the injury/illness estimation server 100 can estimate the possibility that the user 30 will become ill based on blood glucose level information rather than the user's adrenaline secretion amount.

The user terminal 500 is similar to a typical smartphone or tablet terminal, and is similar to a conventional terminal as long as it has at least a communication function and a display function for displaying data, so a detailed description of the configuration will be omitted.

<Summary of the second embodiment>
According to the injury/illness inference system of the second embodiment, the possibility that the user will become ill can be directly inferred from the user's blood glucose level during exercise.

<Additional Information>
The injury/illness inference system according to the above embodiment is not limited to the above embodiment, and each component may be realized by another method. Various modifications will be described below.

(1) In the first embodiment, a learning model that has learned the relationship between blood glucose levels and adrenaline is used to estimate adrenaline from blood glucose levels, but this is not limited to the above. The adrenaline level may be calculated using a predetermined function that inputs blood glucose levels and outputs adrenaline levels. In this case, the adrenaline level may be estimated by weighting the timing at which adrenaline is estimated to be secreted (or the timing at which adrenaline is estimated not to be secreted) for blood glucose level data, which is time-series data.

(2) The configuration of the detection unit 320 shown in the above embodiment is not limited to the configuration shown in FIG. 7. It may be realized with other configurations.

Figure 16 is a schematic diagram of the main components of another detection unit 320 (biomolecular measuring device for other items). The detection unit 320 has a detection element 321 of a specific container shape, which is connected by wiring to a measurement element 700. A bodily fluid sampling section 330 is formed in the detection element 321. The bodily fluid sampling section 330 has a known needle shape, and biological fluids 322 such as sweat and saliva on the skin surface are sampled by surface tension and guided into the detection element 321.

An oxidation-reduction potential measuring electrode 326 and a reference electrode 329 are attached inside the detection element 321. A glucose capture molecule 327 (phenylboronic acid derivative) for capturing glucose 323 as the detection target substance is fixed to the oxidation-reduction potential measuring electrode 326. Since the biological fluid 322 is blood, sweat, saliva, etc., in addition to glucose 323 as the detection target substance, there are also oxidation-reduction potential fluctuation factors 324 such as proteins and garbage, and impurities 325. Note that in the detection unit 320, substances other than glucose may also be detected through the oxidation-reduction potential measuring electrode 326. For this, an antibody or aptamer that captures the detection target substance is used as the capture molecule 327.

In the detection unit 320 (multiple item biomolecule measuring instrument), the measuring element 700 is configured to measure the potential difference between the oxidation-reduction potential measuring electrode 326 of glucose 323 and the reference electrode 329. Then, the amount of glucose is estimated from the change in voltage.
The detection unit 320 may also be realized in the manner shown in FIG.

Figure 17 is a schematic diagram of the main components of another detection unit 320 (biomolecular measuring device for other items). The detection unit 320 has a detection element 321 of a specific container shape, which is connected by wiring to a measurement element 700. A bodily fluid sampling section 330 is formed in the detection element 321. The bodily fluid sampling section 330 has a known needle shape, and biological fluids 322 such as sweat and saliva on the skin surface are sampled by surface tension and guided into the detection element 321.

The detection element 321 is equipped with an oxidation-reduction potential measurement electrode 326, a reference electrode 329, and a counter electrode 331. A glucose capture molecule (phenylboronic acid derivative) for capturing glucose 323 as the detection target substance is fixed to the oxidation-reduction potential measurement electrode 326. Since the biological fluid 322 is sweat, saliva, etc., in addition to glucose 323 as the detection target substance, there are also oxidation-reduction potential fluctuation factors 324 such as proteins and garbage, and impurities 325. Note that the detection unit 320 may also detect substances other than glucose through the oxidation-reduction potential measurement electrode 326. For this, antibodies and aptamers that capture the detection target substance are used as capture molecules.

In the detection unit 320 (biomolecular measuring device for other items), the measuring element 700 is configured to measure the change in the amount of current through the reference electrode 329 and the counter electrode 331 for the oxidation-reduction potential measuring electrode 326 for glucose 323. The amount of glucose is then estimated from the change in current. The detection unit 320 may be configured in this way.

(3) In the above embodiment, an injury/illness estimation system was described that estimates the possibility of a user becoming ill based on the adrenaline secretion amount estimated from the user's blood glucose level during exercise, or based on the blood glucose level. Examples of sensors that measure the blood glucose level are shown in Figures 7, 16, and 17.

The above-mentioned sensor can measure not only blood glucose levels but also redox potential. Therefore, the injury/illness estimation system can also estimate injury/illness using redox potential. In this supplement, we will explain an injury/illness estimation system that can estimate the possibility of injury/illness using the redox potential.

Figure 19 is a conceptual data diagram showing an example of the data configuration of the teacher data in this supplement. As shown in Figure 19, the teacher data 1410 in this supplement is information in which a subject ID 1411, gender 1412, weight 1413, adrenaline data 1414, redox potential data 1801, and disease information 1415 are associated with each other. The teacher data 1410 is a collection of information in which the amount of adrenaline secreted during exercise by the user, the progress of the redox potential during exercise by the user, and disease information indicating an injury or illness suffered by the user are associated with each other. Of this information, information with the same content as the information shown in the above embodiment is assigned the same reference numerals, and detailed description will be omitted.

The redox potential data 1801 in the teacher data 1410 shown in FIG. 19 is information showing the change in redox potential during exercise of the user indicated by the corresponding subject ID 1411. The redox potential can measure whether the state of bodily fluids is in an oxidized or reduced state. Exogenous stress or injury or illness can cause excessive production of free radicals, mainly active oxygen, which can cause bodily fluids to become oxidized. It is known that prolonged periods of hyperlipidemia or hyperglycemia can cause free radicals in blood vessels, which can damage the vascular endothelium and lead to various lifestyle-related diseases such as nephropathy, retinopathy, and periodontal disease (Wells-Knecht, K.J. et al. Biochem. 34, 3702-3709 (1995), etc.). Furthermore, in recent years, there have been reports of blood redox potentials measured in healthy subjects, metabolic syndrome patients, and type 2 diabetes patients (Spanidis, Y. et al. Exp. Therap. Med. 11, 895-903 (2016)), which reveal that the blood redox potential of metabolic syndrome patients and type 2 diabetes patients is higher than that of healthy subjects, and that the blood is in an oxidized state. Therefore, the transition of the redox potential can be used as a reference for estimating injury or illness. The redox potential data 1801 is information measured in the same time series as the corresponding adrenaline data 1414. It is desirable that the timestamps relating to the transition of the changes in the redox potential data 1801 and the adrenaline data 1414 are synchronized with each other, but as long as both are measured during exercise, they may be measured at different times.

By generating a learning model using the teacher data 1410 shown in FIG. 19, this learning model becomes a model that can estimate the possibility that a user will become injured or ill from the transition of the redox potential while the user is exercising and the transition of the amount of adrenaline secreted while the user is exercising. Note that this model can estimate the possibility that a user will become injured or ill by inputting only the transition of the redox potential or the transition of the amount of adrenaline secreted, but it is preferable to input both.

Figure 20 is a sequence diagram showing the procedure for estimating the possibility that a user will become injured or ill using the injury/illness estimation system in this supplement, and is a diagram showing an example of the interaction between each device related to the injury/illness estimation system. Figure 20 shows an example of estimation in a system configured in accordance with the injury/illness estimation system shown in embodiment 1 above.

As shown in FIG. 20, the sensor 300 detects the blood glucose level and the redox potential of the user wearing the sensor 300 during exercise (step S1901). The sensor 300 transmits, among the measured data, pedigree information indicating the change in blood glucose level to the adrenaline estimation server 200 (step S802), and transmits the redox potential information indicating the change in redox potential to the injury/disease estimation server 100 (step S1902). The blood glucose level information and redox potential information transmitted at this time may both include user identification information that can uniquely identify which user the information belongs to, and date and time information indicating when the measurement was performed. This user identification information and date and time information are information that the injury/disease estimation server 100 uses to identify that the redox potential information and the adrenaline secretion amount estimated by the adrenaline estimation server 200 belong to the same user. The assignment of the identification information and date and time information here can also be achieved by assigning the same identification information that is unique in the system to each of the redox potential information and the blood glucose level information.

As shown in the first embodiment above, the adrenaline estimation server 200 estimates the amount of adrenaline secretion from the blood glucose level information, and transmits adrenaline information showing its progress to the injury/illness estimation server 100 (steps S803, S804). This adrenaline information includes information showing the progress of the estimated amount of adrenaline secretion, as well as the user's identification information and date/time information contained in the blood glucose level information.

The injury/illness estimation server 100 receives the redox potential information from the sensor 300. The injury/illness estimation server 100 also receives adrenaline information from the adrenaline estimation server 200. The injury/illness estimation server 100 checks whether the received redox potential information and adrenaline information belong to the same user by comparing the user identification information and date/time information contained in each, and then inputs the redox potential information and adrenaline information into the learning model. Then, based on this input, the estimation unit 132 of the injury/illness estimation server 100 estimates the possibility that the user will suffer an injury or illness (step S1903). The injury/illness estimation server 100 then transmits injury/illness estimation information indicating the estimation result to the user terminal 500 (step S806).

When the user terminal 500 receives the injury/illness estimation information, it displays the contents (step S807). This allows the user to recognize the degree to which the user is likely to become injured or ill, and the type of injury or illness that the user is likely to suffer, based on the trends in the redox potential and the trends in the amount of adrenaline secreted during exercise.

In addition, the injury/illness estimation system in this supplement learns information including the changes in adrenaline secretion during exercise and the changes in redox potential as teacher data, and is therefore able to estimate the possibility of an injury or illness based on the changes in adrenaline secretion during exercise and the changes in redox potential, but estimation using the changes in redox potential is not limited to this. Needless to say, examples other than those shown in this supplement may also be used.

For example, in the training data shown in FIG. 18, the change in blood glucose level may be used instead of the change in adrenaline secretion. In this case, an injury/illness estimation system can be provided that accepts input of the change in blood glucose level and the change in redox potential during exercise of a user, and outputs information indicating the possibility that the user may become injured or ill.

In addition, in the above-mentioned first and second embodiments, the possibility of the user becoming ill may be estimated based on the transition of the redox potential during the user's exercise, separately from the estimation of the possibility of the user becoming ill based on the transition of the adrenaline secretion amount and the transition of the blood glucose level. That is, the injury/illness estimation system may estimate the possibility of the user becoming ill based on the transition of the adrenaline secretion amount during the user's exercise, estimate the possibility of the user becoming ill based on the transition of the redox potential during the user's exercise, and output both estimation results. Also, instead of outputting both estimation results, information on the injury or illness that is estimated to be likely to cause the same type of injury or illness in both estimation results may be output. Also, information on the injury or illness that is estimated to be likely to cause the injury or illness in both estimation results and for which the possibility of the injury or illness is equal to or greater than a predetermined threshold may be output.

As shown in this supplement, by estimating the possibility of a user becoming ill using the change in the redox potential during exercise as information other than blood glucose level or the amount of adrenaline secreted based on blood glucose level, it is possible to estimate the possibility of becoming ill that cannot be estimated from blood glucose level or the amount of adrenaline secreted alone, and to increase the accuracy of the possibility of becoming ill that is estimated from the user's blood glucose level or the amount of adrenaline secreted during exercise.

(4) In the above embodiment, the injury/illness estimation system estimates the possibility of contracting an injury or illness by having the processor of the injury/illness estimation server execute an injury/illness estimation program or the like, but this may also be realized by a logic circuit (hardware) or dedicated circuit formed in an integrated circuit (IC (Integrated Circuit) chip, LSI (Large Scale Integration)) or the like in the device. Furthermore, these circuits may be realized by one or more integrated circuits, and the functions of the multiple functional units shown in the above embodiment may be realized by a single integrated circuit. LSIs are sometimes called VLSIs, super LSIs, ultra LSIs, etc. depending on the degree of integration.

The injury/illness estimation program may be recorded on a processor-readable recording medium, and the recording medium may be a "non-transient tangible medium" such as a tape, disk, card, semiconductor memory, or programmable logic circuit. The injury/illness estimation program may be supplied to the processor via any transmission medium (such as a communication network or broadcast waves) capable of transmitting the injury/illness estimation program. In other words, the injury/illness estimation program may be downloaded from a network and executed using an information processing device such as a smartphone. The present invention may also be realized in the form of a data signal embedded in a carrier wave, in which the injury/illness estimation program is embodied by electronic transmission.

The injury/illness estimation program can be implemented using, for example, a scripting language such as ActionScript or JavaScript (registered trademark), or an object-oriented programming language such as Objective-C, Java (registered trademark), C++, Python, or R.

(4) The various configurations described in the above embodiments and supplementary notes may be combined as appropriate.

100 Injury/disease estimation server 110 Communication unit 120 Input unit 130 Control unit 131 Reception unit 132 Estimation unit 133 Learning unit 140 Memory unit 141 Injury/disease estimation learning model 150 Output unit 200 Adrenaline estimation server 210 Communication unit 220 Input unit 230 Control unit 231 Reception unit 232 Estimation unit 233 Learning unit 240 Memory unit 241 Adrenaline estimation learning model 250 Output unit 300 Sensor 310 Communication unit 320 Detection unit 340 Memory unit 341 Sensing data 350 Output unit

Claims (6)

A reception unit that receives input of blood glucose level information regarding a blood glucose level of a target user during exercise;
A storage unit that stores an adrenaline estimation learning model that learns the relationship between a blood glucose level of a user during exercise and an adrenaline secretion amount, and an injury/illness estimation learning model that learns the relationship between the adrenaline secretion amount of the user during exercise and the injury/illness of the user ;
an adrenaline secretion amount estimation unit that estimates an adrenaline secretion amount of the target user by using the blood glucose level information received by the reception unit and the adrenaline estimation learning model;
An injury/illness estimation unit that estimates the possibility that the target user will suffer from an injury or illness using the estimated adrenaline secretion amount and the injury/illness estimation learning model;
an output unit that outputs information regarding the possibility estimated by the estimation unit;
An injury/illness estimation system equipped with The injury/ illness estimation learning model is a learning model that learns a relationship between a change in the adrenaline secretion amount of the user and the injury/illness of the user,
The injury/illness inference system according to claim 1 , wherein the receiving unit receives an input of blood glucose level information relating to a change in blood glucose level of the target user during exercise. The injury/ illness estimation learning model is a model that further learns a relationship between an oxidation-reduction potential of the user during exercise and the injury/illness of the user,
The receiving unit further receives an input of oxidation-reduction potential information regarding an oxidation-reduction potential of the target user during exercise,
The injury/illness inference system according to claim 1 or 2, characterized in that the injury/illness inference unit estimates the possibility that the target user will suffer from an injury or illness using the blood glucose level information and the oxidation-reduction potential information accepted by the acceptance unit and the learning model. 4. The injury/illness estimation system according to claim 1, wherein the injury/illness is any one of obesity, diabetes, and renal disorder . The computer
A receiving step of receiving input of blood glucose level information regarding the blood glucose level of the target user during exercise;
a storage step of storing an adrenaline estimation learning model that has learned a relationship between a blood glucose level of a user during exercise and an adrenaline secretion amount, and an injury/illness estimation learning model that has learned a relationship between the adrenaline secretion amount of the user during exercise and the injury/illness of the user ;
an adrenaline secretion amount estimation step of estimating an adrenaline secretion amount of the target user using the blood glucose level information received in the receiving step and the adrenaline estimation learning model;
an injury/illness estimation step of estimating a possibility that the target user will suffer from an injury or illness using the estimated adrenaline secretion amount and the injury/illness estimation learning model;
an output step of outputting information regarding the possibility estimated in the estimation step;
An injury estimation method that performs the above. On the computer,
A reception function for receiving input of blood glucose level information regarding the blood glucose level of the target user during exercise;
A memory function for storing an adrenaline estimation learning model that learns the relationship between the blood glucose level of a user during exercise and the amount of adrenaline secreted, and an injury/illness estimation learning model that learns the relationship between the amount of adrenaline secreted by the user during exercise and the injury/illness of the user ;
an adrenaline secretion amount estimation function that estimates the adrenaline secretion amount of the target user by using the blood glucose level information received by the reception function and the adrenaline estimation learning model;
An injury/illness estimation function that estimates the possibility that the target user will suffer from an injury or illness using the estimated adrenaline secretion amount and the injury/illness estimation learning model;
an output function for outputting information regarding the possibility estimated by the estimation function;
An injury/illness estimation program that makes this possible.

Images