Living body symptom change prediction system, information processing apparatus, and living body symptom change prediction method

Abstract

To predict a change in a symptom of a living body caused by environmental factors by using a smaller amount of data.SOLUTION: The present disclosure provides a living body symptom change predicting system 100 having an environment sensor 11 and an information processing device 60. The environment sensor detects environmental factor data relating to environmental factors in a target space. The information processing device has a control unit for converting the environmental factor data into intermediate data by means of a model that simulates biological characteristics associated with symptoms relating to the environment. The control unit uses associating information that associates at least the intermediate data and the possibility of a symptom change relating to the environment to predict the possibility of the symptom change from the intermediate data.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a biological symptom change prediction system, an information processing apparatus, and a biological symptom change prediction method.

Background Art

[0002] It is known that changes occur in human symptoms such as the onset of asthma and allergic symptoms due to environmental factors (e.g., temperature, humidity, CO2, PM2.5, TVOC, formaldehyde, etc.). If it were possible to predict human symptom changes to some extent from environmental factor data, it would be possible for patients to avoid that environment or to perform environmental control in advance so as to create an environment in which human symptoms improve.

[0003] Techniques are known for detecting events that may exacerbate chronic diseases and recommending actions etc. based on the detection results (see, for example, Patent Document 1). Patent Document 1 discloses a technique for detecting events that may exacerbate chronic diseases from physiological data and environmental factor data and recommending preferable actions and medications based on the detection results.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in order to create response information for predicting biological symptom changes from environmental factor data by means of machine learning or the like, an enormous amount of data is required. This is because when a person is affected by the environment, there are some internal reactions or changes, and as a result, symptoms occur, and there is rarely a simple linear relationship between the symptoms and the environment.

[0006] The present disclosure provides a technique for predicting changes in symptoms of a living body caused by environmental factors with less data.

Means for Solving the Problem

[0007] The first aspect of the present disclosure is a symptom change prediction system for a living body having an environmental sensor and an information processing device, wherein the environmental sensor detects environmental factor data regarding environmental factors in a target space, the information processing device has a control unit that converts the environmental factor data into intermediate data by a model simulating biological characteristics related to symptoms with respect to the environment, the control unit predicts the possibility of symptom change from the intermediate data using correspondence information associating at least the intermediate data with the possibility of symptom change with respect to the environment.

[0008] According to the first aspect of the present disclosure, it is possible to predict changes in human symptoms caused by environmental factors with less data.

[0009] The second aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics is Weber-Fechner's law.

[0010] The third aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics is a model that outputs different intermediate data depending on the range of values taken by the environmental factor data or whether the environmental factor data satisfies a condition.

[0011] The fourth aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics is a model that outputs the intermediate data corresponding to the amount of increase or decrease only when the change in the environmental factor data changes either to increase or decrease, or a model that outputs the intermediate data that changes according to the value of the nth derivative.

[0012] A fifth aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics outputs the intermediate data according to the integrated value of the environmental factor data, or the intermediate data that is different depending on the change history of the past values of the environmental factor data even if the current value of the environmental factor data is the same.

[0013] A sixth aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics outputs the intermediate data according to the duration for which the environmental factor data continues to have values within a predetermined range, or the number of times the environmental factor data repeats values within a predetermined range.

[0014] A seventh aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics outputs the intermediate data that changes after a predetermined time has elapsed since the time when the environmental factor data occurred.

[0015] An eighth aspect of the present disclosure is the symptom change prediction system according to the first aspect, wherein the model simulating the biological characteristics is a log function, an exponential function, or an nth-order function that takes the environmental factor data as input and outputs the intermediate data.

[0016] A ninth aspect of the present disclosure is the symptom change prediction system according to any one of the first to eighth aspects, wherein the control unit further predicts the possibility of the symptom change from the intermediate data and the environmental factor data using the correspondence information associated with the environmental factor data.

[0017] A tenth aspect of the present disclosure is the symptom change prediction system according to any one of the first to ninth aspects, wherein The possibility of the symptom change is that any one of allergic symptoms, asthmatic symptoms, meteoropathy, infectious diseases, deteriorated sleep quality, reduced wakefulness and drowsiness, autonomic nerve disorder, frailty, dementia and delirium, reduced memory, reduced motor function, heat stroke, motion sickness, or VR sickness may deteriorate or improve. The 11th aspect of the present disclosure is the symptom change prediction system according to any one of the 1st to 10th aspects, The environmental factor data is a statistic processed by statistical processing.

[0018] The 12th aspect of the present disclosure is the symptom change prediction system according to the 9th aspect, The control unit predicts the possibility of the symptom change from the actually measured environmental factor data and the intermediate data, or from the forecast value of the environmental factor data and the intermediate data.

[0019] The 13th aspect of the present disclosure is the symptom change prediction system according to any one of the 1st to 12th aspects, Using the symptom change information reported for asthmatic symptoms, allergic symptoms, meteoropathy, infectious diseases, deteriorated sleep quality, reduced wakefulness and drowsiness, autonomic nerve disorder, frailty, dementia and delirium, reduced memory, reduced motor function, heat stroke, motion sickness, or VR sickness as teacher data, the correspondence information is generated by a machine learning method.

[0020] The 14th aspect of the present disclosure is the symptom change prediction system according to the 13th aspect, The control unit generates first correspondence information using the intermediate data as an explanatory variable and the symptom change information reported by a plurality of people as teacher data by a machine learning method, The control unit generates second correspondence information using the intermediate data as an explanatory variable and the symptom change information reported by an individual as teacher data by a machine learning method, The control unit predicts the possibility of an individual's symptom change based on the possibilities of symptom change predicted by the first correspondence information and the second correspondence information, respectively.

[0021] A 15th aspect of the present disclosure is the symptom change prediction system according to the 13th or 14th aspect, wherein the control unit generates the correspondence information for each season, for each target disease, or for each allergen in allergic symptoms, and predicts the possibility of symptom change based on the correspondence information generated for each season, for each target disease, or for each allergen.

[0022] A 16th aspect of the present disclosure is the symptom change prediction system according to any one of the 1st to 15th aspects, wherein the control unit displays the intermediate data and the possibility of symptom change on the same screen.

[0023] A 17th aspect of the present disclosure is an information processing apparatus, comprising receiving environmental factor data regarding environmental factors in a target space from the environmental sensor, having a control unit that converts the environmental factor data into intermediate data by a model that simulates biological characteristics related to symptoms for the environment, wherein the control unit predicts the possibility of symptom change from the intermediate data using correspondence information that associates at least the intermediate data with the possibility of symptom change for the environment.

[0024] According to the 17th aspect of the present disclosure, it is possible to predict a change in a biological symptom caused by environmental factors with less data.

[0025] A 18th aspect of the present disclosure is a symptom change prediction method performed by a symptom change prediction system for a living body having an environmental sensor and an information processing apparatus, comprising the environmental sensor detecting environmental factor data regarding environmental factors in a target space, the control unit performing a process of converting the environmental factor data into intermediate data by a model that simulates biological characteristics related to symptoms for the environment, and performing a process of predicting the possibility of symptom change from the intermediate data using correspondence information that associates at least the intermediate data with the possibility of symptom change.

[0026] According to the 18th aspect of the present disclosure, it is possible to predict changes in the symptoms of a living body caused by environmental factors with less data.

Brief Description of the Drawings

[0027]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Embodiments for Carrying Out the Invention

[0028] Hereinafter, as an example of embodiments for carrying out the present disclosure, a human symptom change prediction system and a human symptom change prediction method performed by the human symptom change prediction system will be described.

[0029] <Regarding Intermediate Data and Human Symptom Changes> Symptoms that occur in humans are affected by environmental factors. Symptoms include allergic symptoms, asthma symptoms, meteorological diseases, infectious diseases, reduced sleep quality (insomnia, waking up during sleep, difficulty falling asleep, poor waking up), reduced wakefulness and sleepiness, autonomic nervous system disorders, frailty, dementia and delirium, memory loss, motor function decline, heat stroke, motion sickness, VR sickness, etc. Thus, symptoms are not limited to diseases.

[0030] However, there is not a simple linear relationship between the symptoms that occur in humans and environmental factors. It cannot be regarded as a simple linear relationship because there are some reactions or changes in the human body under the influence of environmental factors, and as a result, symptoms occur. The relationship between environmental factors and symptoms is known to be a non-linear relationship and is very complex. Therefore, a huge amount of data is required to create a model from the relationship between the two using machine learning or the like.

[0031] Therefore, in the present embodiment, by using intermediate data indicating the reaction (symptoms) of the human body to environmental factors as an explanatory variable, it is possible to accurately predict symptoms with a smaller amount of data. Intermediate data is a numerical value that simulates the response related to environmental factors that cause symptoms. Such intermediate data can be generated by a model that simulates biological characteristics (described later). The model itself for creating intermediate data can simulate symptom characteristics. By converting environmental factor data into intermediate data using a model that simulates biological characteristics, the accuracy of machine learning can be improved with a small number of samples.

[0032] In addition, in this embodiment, this intermediate data may be described using the term "sensory intensity". Also, changes in symptoms may be described using the term "worsening risk".

[0033] As specific environmental factors, environmental factors related to air quality such as indoors, in a vehicle, or outdoors (current status: temperature, humidity, CO2, PM2.5, TVOC (total volatile organic compounds), formaldehyde, etc.) are known. Due to changes in environmental factors, the risk of exacerbation of asthma or allergy symptoms may increase. Exacerbation refers to a state where diseases such as asthma or allergy symptoms deteriorate, cannot be improved by normal treatment, and the treatment content needs to be changed. The worsening risk refers to the risk (possibility) of such a state.

[0034] For example, it is known that there is a correlation between asthma and environmental factors. Asthma is a symptom said to be contributed to by a very wide range of environmental factors in its onset. As environmental factors that exacerbate asthma, PM2.5, VOC, SOx, NOx, fungi, mites, dust, strong odors / smoke, smoke, cold and dry air, hot air, etc. are known, and they are diverse.

[0035] The Ministry of Health, Labour and Welfare has published reference values and recommended values for preferable environmental factors not only for the suppression of asthma.

[0036] Temperature (17°C to 28°C), humidity (40% to 70%), carbon dioxide concentration (1000 ppm or less), formaldehyde (0.08 ppm or less) However, these reference values and recommended values are a guideline for the presence or absence of health risks to the human body, and there are multiple factors contributing to symptoms.

[0037] Therefore, according to experts, there is currently no common countermeasure for everyone against asthma or allergy symptoms, and it is also difficult to identify the cause. There are also patients who do not show symptoms during diagnosis or when admitted for examination, but only show symptoms at home. Also, self-medication at home is said to be important.

[0038] Based on these circumstances, environmental factors that reduce asthma or allergy symptoms cannot be fully addressed by setting reference values for each factor. It is also important that the patient himself / herself can approach the indoor environment, and the operability of environmental control is also considered important.

[0039] <Outline of the method for predicting exacerbation risk> FIG. 1 is a diagram for explaining the system configuration and the outline of the prediction method of an example of a human symptom change prediction system 100. The environmental device 10, the environmental sensor 11, the user terminal 70, and the information processing device 60 are communicably connected via the network N.

[0040] The environmental sensor 11 is preferably installed in each room of the building where the patient 9 having asthma or allergy symptoms lives. The environmental sensor 11 detects environmental factor data regarding the environmental factors of the target space 7 where the patient 9 lives. The environmental sensor 11 may be installed only once in the building. The environmental device 10 is a device that controls the environment related to air quality, such as an air conditioner, a ventilation device, or an air purifier. The environmental device 10 may have a plurality of functions, or there may be environmental devices 10 for each function.

[0041] The user terminal 70 is a terminal device that displays a map screen to be described later. In the user terminal 70, a web browser or a native app is executed, and information for display on the display is received from the information processing device 60 via the network N. The patient 9 can browse the map screen or the like to grasp the environmental situation that reduces or does not increase the exacerbation risk. The user terminal 70 can be carried by the patient 9 and does not have to be installed in the same space as the space where the environmental sensor 11 is installed.

[0042] The information processing device 60 is, for example, a server device that performs various information processing, provides services, and stores files. The information processing device 60 generates a mathematical model to be described later, and predicts the exacerbation risk by inputting the sensory intensity and environmental factor data into this mathematical model. The sensory intensity is a variable representing the magnitude of the influence that a change in the air quality environment will have on a human being as a stimulus.

[0043] (1) The environmental sensor 11 transmits environmental factor data to the information processing device 60. In the learning phase for generating a mathematical model, further, the patient 9 self-reports symptom change information (e.g., deteriorated, improved, remitted, ameliorated) from the user terminal 70 to the information processing device 60.

[0044] (2) The information processing device 60 inputs the generated sensory intensity and environmental factor data into the mathematical model, and predicts the exacerbation risk. This mathematical model uses the sensory intensity without requiring vital data as described later. The mathematical model is correspondence information associating the sensory intensity with the exacerbation risk of asthma and allergic symptoms.

[0045] (3) The information processing device 60 transmits to the user terminal 70 a map screen on which the exacerbation risk is arranged with respect to the environmental factor data.

[0046] (4) The user terminal 70 displays the map screen, and the map screen shows what environmental settings will reduce or increase the exacerbation risk. By specifying the displayed environmental settings, the user terminal 70 transmits the environmental settings to the information processing device 60.

[0047] (5) The information processing device 60 converts the environmental settings into the setting information of the environmental device 10 and transmits it to the environmental device 10. Thereby, the environmental device 10 can control the air quality so as to reduce the exacerbation risk or not increase it.

[0048] Thus, in the present disclosure, vital data is not used in predicting the exacerbation risk. Specifically, in addition to environmental factor data, "intermediate data (sensation intensity)" is introduced as a variable representing the magnitude of the impact that humans will receive, with the change in the air quality environment acting as a stimulus (vital data can be substituted), enabling the exacerbation risk of symptoms to be predicted with high accuracy. Also, by converting environmental factor data into intermediate data using a model that simulates biological characteristics, the accuracy of machine learning can be improved with a small number of samples. This makes it possible for patient 9 to avoid that environment or perform environmental control to an indoor environment with a reduced exacerbation risk.

[0049] <<Modified Example of System Configuration>> Rather than the environmental sensor 11 existing independently, as shown in FIG. 2, the environmental sensor 11 may be built into the indoor unit 10b. FIG. 2 shows a modified example of the system configuration of the human symptom change prediction system 100. The human symptom change prediction system 100 mainly includes one outdoor unit 10a as a heat source unit, one or more indoor units 10b as utilization units, and a remote control device (hereinafter referred to as the "remote control 12") as an input device for inputting commands related to various settings.

[0050] The outdoor unit 10a and the indoor unit 10b are referred to as an air conditioner. The outdoor unit 10a and the indoor unit 10b are connected by a refrigerant connection pipe (gas connection pipe GP) to form a refrigerant circuit. Also, in the human symptom change prediction system 100, a plurality of communication networks (network NW1, network NW2) that function as signal transmission paths between units are constructed. The network NW2 may be wired or wireless.

[0051] The remote control 12 is a user interface that accepts settings such as temperature and humidity. The function of the information processing device 60 may be the same as that in FIG. 1.

[0052] (1) The indoor unit 10b has a built-in integrated environmental sensor 11. The indoor unit 10b transmits environmental factor data to the outdoor unit 10a.

[0053] (2) The outdoor unit 10a transmits environmental factor data to the information processing device 60.

[0054] (3) The information processing device 60 inputs the generated sensory intensity and environmental factor data into a mathematical model, and predicts the exacerbation risk.

[0055] (4) The information processing device 60 transmits a map screen on which the exacerbation risk is arranged for the environmental factor data to the remote controller 12 via the outdoor unit 10a and the indoor unit 10b.

[0056] (5) The remote controller 12 displays the map screen, and the map screen shows what environmental settings will reduce or increase the exacerbation risk. By specifying the displayed environmental settings, the remote controller 12 transmits the environmental settings to the indoor unit 10b.

[0057] (6) The indoor unit 10b converts the environmental settings into the setting information of the air conditioner and controls itself. Thereby, the environmental device 10 can control the air quality so as to reduce the exacerbation risk or not increase it.

[0058] <System Configuration of Human Symptom Change Prediction System> Next, with reference to FIG. 3, the system configuration of the human symptom change prediction system 100 will be described. FIG. 3 is a diagram showing an example of the system configuration of the human symptom change prediction system 100.

[0059] The human symptom change prediction system 100 provides various services utilizing IoT from administrators to general users by communicating various devices 30 such as air conditioners and lighting and the information processing device 60 on the cloud side via the network N. The edge device 80, the devices 30, the sensor switches 53, and the user terminal 70 are installed on the customer side, and the information processing device 60 is installed in the cloud such as a data center or the Internet. Note that since the edge device 80 is a device for centrally managing the devices 30 and the sensor switches 53, the edge device 80 may not be provided.

[0060] The device 30 refers to all devices that consume power, such as environmental devices 10, security facilities, heat source devices, fire alarms, AHU (Air Handling Unit), electricity meters, lighting, etc. The sensor switches 53 include various sensors such as environmental sensor 11, lamps, relays, etc. The device 30 and the sensor switches 53 are communicably connected to the edge device 80 via a dedicated cable or a network such as a LAN. The device 30 and the sensor switches 53 may be communicably connected to the edge device 80 by wireless communication.

[0061] The device 30 and the sensor switches 53 are controlled by the edge device 80. In other words, the edge device 80 performs the required operations on the device 30 and the sensor switches 53 so as to conform to the purposes of the device 30 and the sensor switches 53. The content of the control varies depending on the types of the device 30 and the sensor switches 53. For example, when the device 30 is an air conditioner, all controls related to the functions of the air conditioner, such as the cooling / heating mode, set temperature, air volume, humidity, air direction, etc., which can generally be set on the air conditioner, may be included. Also, the control includes operation modes such as a pre-season inspection dedicated mode, microcomputer reset, operation stop, and function substitution.

[0062] The device 30 collects operation data corresponding to the device 30 and mainly transmits it to the edge device 80 periodically. Periodically means, for example, once per minute, once per 10 minutes, once per 60 minutes, etc., but it may be set by the user or the information processing device 60. Also, upon request from the edge device 80 or the user terminal 70, the device 30 can transmit the operation data to the edge device 80. The operation data varies depending on the device 30. For example, in the case of an air conditioner, it includes the high-pressure of the refrigerant, low-pressure of the refrigerant, refrigerant temperature, rotation speed of the fan, and CPU temperature of the microcomputer, etc.

[0063] Also, when the device 30 detects an abnormality, it transmits an abnormality code to the edge device 80. The device 30 that has detected the abnormality stops operating. The edge device 80 transmits the abnormality code to the information processing device 60. The same processing by the edge device 80 may be applied to the sensor switches 53. The sensor switches 53 transmit various detection information such as the detected environmental factor data to the edge device 80, for example, periodically, or transmit an abnormality code.

[0064] The edge device 80 is a controller that controls the device 30 and the sensor switches 53. In the absence of the edge device 80, the information processing device 60 controls the device 30 and the sensor switches 53. The edge device 80 has functions as a control device that controls the device 30 and the sensor switches 53, an information processing device that processes operation data and the like, and a communication device that communicates with the information processing device 60. For example, the edge device 80 transmits the abnormality code received from the device 30 to the information processing device 60 and receives an instruction corresponding to the abnormality code from the information processing device 60. Alternatively, even if the edge device 80 does not transmit any information to the information processing device 60, it receives an instruction from the information processing device 60 (for example, when there is an instruction from the user terminal 70 to the information processing device 60). The edge device 80 converts the instruction into an appropriate instruction according to the models of the device 30 and the sensor switches 53 and transmits it to the device 30 and the sensor switches 53.

[0065] The information processing device 60 may be one or more server devices. Although one information processing device 60 is shown in FIG. 3, the information processing device 60 may be divided and installed into several parts according to functions. Also, the functions of the information processing device 60 may be aggregated by one server device. Also, a plurality of information processing devices 60 with the same functions may be prepared, and a plurality of information processing devices 60 may process while communicating like a server cluster.

[0066] The information processing device 60 inputs the sensory intensity and environmental factor data into a mathematical model to predict the exacerbation risk, and provides it to the user terminal 70 or the like. The information processing device 60 can predict the exacerbation risk not only for individual facilities such as houses and buildings, but also for each region, and may share the exacerbation risk with public broadcasting or the like. In addition, the information processing device 60 can use the comprehensively prepared environmental factor data and exacerbation risk to provide the user terminal 70 with a preferable environmental setting for the exacerbation risk. Further, the information processing device 60 can also transmit an instruction to the device 30 to the edge device 80 according to the schedule and operation set by the user terminal 70.

[0067] Although not shown in FIG. 3, an information processing device that generates a mathematical model may exist separately from the information processing device 60. In this case, the mathematical model generated by the information processing device is introduced into the information processing device 60. In the present disclosure, for convenience of explanation, it is assumed that the information processing device 60 generates a mathematical model.

[0068] The information processing device 60 also has the function of a Web server. The Web server responds to requests from client software (Web client) such as a Web browser operated by the user, and provides the client with screen information described in an HTML file, XML, CSS file, JavaScript (registered trademark), or the like. An application that uses the Web mechanism in this way is called a Web application.

[0069] Note that the information processing device 60 preferably supports cloud computing. Cloud computing refers to a usage form in which resources on a network are used without awareness of specific hardware resources. Conventionally, cloud computing provides data and software that were conventionally used by the user on their local computer to the user as a service via a network. On the user side, by preparing a Web browser operating on a personal computer, a mobile information terminal, or the like, and an Internet connection environment, various services can be used from any terminal.

[0070] The user terminal 70 is a client terminal that displays various screens provided by the information processing device 60. The user terminal 70 may be used by an administrator or a general user (patient 9 in the present disclosure). When the device 30 is in a general household, the administrator may be a family member, or the patient 9 may also serve as the administrator. When the device 30 is in a building or the like managed by a company, the administrator may be, for example, a facility administrator, a caregiver, a nurse, or the like.

[0071] Although the screens displayed by the user terminal 70 are diverse, as an example, there are the above-mentioned map screen, a list screen of the devices 30 and sensor switches 53 existing on the customer side, an in-house map showing the locations where the devices 30 and sensor switches 53 are arranged, and an operation screen for operating the devices 30 and sensor switches 53.

[0072] The user terminal 70 is, for example, a PC (Personal Computer), a smartphone, a tablet terminal, a PDA (Personal Digital Assistant), a wearable PC (such as a sunglass type or a wristwatch type), etc. However, it is only necessary to have a communication function and for the web browser to operate. Also, instead of a web browser on the user terminal 70, a native app dedicated to the human symptom change prediction system 100 may operate.

[0073] <Hardware Configuration of Information Processing Device> Referring to FIG. 4, the hardware configuration of the information processing device 60 will be described. FIG. 4 is a diagram showing an example of the hardware configuration of the information processing device 60. As shown in FIG. 4, the information processing device 60 includes a processor 221, a memory 222, an auxiliary storage device 223, an I / F (Interface) device 224, a communication device 225, and a drive device 226. Each hardware of the information processing device 60 is interconnected via a bus 207.

[0074] The processor 221 has various arithmetic devices such as a CPU (Central Processing Unit). The processor 221 reads out various programs onto the memory 222 and executes them. The processor 211 corresponds to a control unit 110 that controls and performs arithmetic operations on the entire information processing apparatus 60. In addition to overall control, the control unit 110 performs a process of estimating the sensory intensity related to human senses. The process of estimating the sensory intensity will be described later.

[0075] The memory 222 has main memory devices such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The processor 221 and the memory 222 form a so-called computer, and the processor 221 executes various programs read onto the memory 222.

[0076] The auxiliary storage device 223 stores various programs and various data used when the various programs are executed by the processor 221.

[0077] The I / F device 224 is a connection device that connects a display device 230 and an operation device 240, which are examples of external devices, to the information processing apparatus 60. The display device 230 displays the internal state of the information processing apparatus 60. The operation device 240 is used when an administrator of the information processing apparatus 60 inputs various instructions to the information processing apparatus 60.

[0078] The communication device 225 is a communication device for communicating with the edge device 80 and the user terminal 70 via the network N.

[0079] The drive device 226 is a device for setting the recording medium 250. The recording medium 250 here includes media that optically, electrically, or magnetically record information, such as a CD-ROM, a flexible disk, and a magneto-optical disk. Further, the recording medium 250 may include semiconductor memories that electrically record information, such as a ROM and a flash memory.

[0080] Note that various programs installed in the auxiliary storage device 223 are installed, for example, when the distributed recording medium 250 is set in the drive device 226 and various programs recorded on the recording medium 250 are read by the drive device 226. Alternatively, various programs installed in the auxiliary storage device 223 may be installed by being downloaded from the network N via the communication device 225.

[0081] <Regarding functions> Next, with reference to FIG. 5, the functional configurations of the respective devices included in the human symptom change prediction system 100 will be described in detail. FIG. 5 is an example of a functional block diagram for explaining, in a block-divided manner, the functions of the information processing device 60 in the learning phase.

[0082] The information processing device 60 includes an environmental factor data acquisition unit 61, a symptom change information reception unit 62, a statistic calculation unit 63, a sensory intensity estimation unit 64, and a mathematical model generation unit 65. Each of these units included in the information processing device 60 is a function or means realized by the control unit 110 of the information processing device 60 executing the instructions of the program expanded in the memory 222.

[0083] The environmental factor data acquisition unit 61 acquires environmental factor data representing, for example, the concentration of environmental factors actually detected by the environmental sensor 11. In the present disclosure, the environmental factor data is, for example, temperature, humidity, CO2 concentration, PM2.5 concentration, TVOC concentration, and formaldehyde concentration, but this is merely an example. For example, pollen or dust may be included in the environmental factor data.

[0084] The environmental factor data acquisition unit 61 may simply receive the environmental factor data periodically transmitted by the environmental sensor 11, or may request the environmental sensor 11 to measure the environmental factor data and receive the environmental factor data as a response thereto. The acquisition timing is preferably at least once a day or more, and for example, it is conceivable to set the frequency to once every few minutes to several hours.

[0085] The symptom change information reception unit 62 receives symptom change information input by the patient 9 with a high risk of exacerbation on the user terminal 70. The symptom change information is information indicating the presence or absence of symptoms such as asthma and allergic symptoms, and the intensity of the symptoms. For example, the patient 9 inputs "1" when symptoms appear and "0" when symptoms do not appear as the symptom change information for that day every day. The symptom change information may be a numerical value from 0 to 100 or an intensity in 3 to 5 levels instead of "1" or "0". In this way, the symptom change information represents not only deterioration but also improvement. These numerical values correspond to the symptoms getting worse, improving, remitting, or being alleviated. When the patient 9 logs in to the information processing device 60, the symptom change information reception unit 62 acquires the patient ID (able to identify the patient). The basic information about the patient 9 is registered in the information processing device 60 in advance. For example, the basic information includes age, gender, height, allergens (such as mold, dust, pollen), underlying diseases, etc. An allergen refers to an antigen that specifically reacts with the antibodies of a person with asthma or allergic diseases. Generally, an allergen refers to a substance that causes the allergic symptoms.

[0086] The statistic calculation unit 63 calculates the statistic of the environmental factor data by processing the environmental factor data through statistical processing. The statistic may be, for example, the maximum value per day, the minimum value per day, the average value of a day, the standard deviation, etc. for each environmental factor data. Also, the statistic is not limited to these and may be the median, the integrated value, the moving average from the past few hours to several days, etc. The reason for calculating the statistic is to reduce the processing load of the environmental factor data, so the statistic calculation unit 63 may not be provided.

[0087] The sensory intensity estimation unit 64 generates intermediate data based on a model that simulates biological characteristics from the statistic of the environmental factor data. As an example in the present disclosure, the sensory intensity estimation unit 64 estimates the sensory intensity from the statistic of the environmental factor data. The details of the sensory intensity will be described with reference to FIGS. 19 and 20.

[0088] The mathematical model generation unit 65 generates a mathematical model for predicting the exacerbation risk from the statistical quantity of environmental factor data and the sensory intensity, using learning data with the statistical quantity of environmental factor data and the sensory intensity as explanatory variables and the symptom change information as the objective variable (teacher data). A mathematical model is a simplified representation of a real-world object that mathematically expresses the relationships between various quantities according to physical laws. However, it is not required to be a rigorous mathematical model, and it is sufficient if the mathematical model can predict the exacerbation risk from the environmental factor data and the sensory intensity.

[0089] <Example of environmental factor data> Referring to FIG. 6, the environmental factor data that can be detected by the environmental sensor 11 will be described. FIG. 6 shows the environmental factor data acquired by the environmental factor data acquisition unit 61. Here, for the sake of convenience of explanation, FIG. 6 shows the statistical quantity of the environmental factor data. As shown in FIG. 6, the statistical quantity of temperature (daily average value, standard deviation, maximum value per day, minimum value per day), the statistical quantity of humidity, (omitted), the statistical quantity of PM2.5 concentration, etc. have been acquired. The environmental factor data in FIG. 6 is the statistical quantity per day, but the statistical quantity for a shorter time may also be used.

[0090] Also, FIG. 6 shows the symptom change information corresponding to the environmental factor data. The symptom change information is input by the patient 9. This symptom change information is either exacerbated (1) or not exacerbated (0).

[0091] <Generation of mathematical model> Subsequently, referring to FIGS. 7 to 9, the method for generating the mathematical model will be described. FIG. 7 is a diagram for explaining the learning phase of generating a mathematical model from the statistical quantity of environmental factor data, the sensory intensity (explanatory variable), and the symptom change information (objective variable, teacher data). The functional configurations of FIGS. 5 and 7 are the same, but in FIG. 7, the blocks are arranged along the processing flow.

[0092] S1: First, the environmental factor data acquisition unit 61 acquires the environmental factor data actually detected by the environmental sensor 11. The environmental factor data is not limited to the measured value and may also be a forecast value. This forecast value may be provided by the Japan Meteorological Agency or the like, or may be predicted by the environmental factor data acquisition unit 61 from the measured value.

[0093] S2: Further, the symptom change information input from a patient 9 having asthma or allergy symptoms from the user terminal 70 or the like is received by the symptom change information receiving unit 62.

[0094] S3: Next, the statistic calculation unit 63 calculates the statistic of the environmental factor data (for example, the maximum value per day, the minimum value per day, the average value of a day, the standard deviation, etc.). Thus, the environmental factor data shown in FIG. 6 is obtained.

[0095] S4: Next, the sensation intensity estimation unit 64 estimates the sensation intensity from the statistic of the environmental factor data. When a forecast value is used for the environmental factor data, the sensation intensity is also estimated using the forecast value. Details of the sensation intensity will be described with reference to FIGS. 19 and 20. The sensation intensity P is estimated by Equation (1).

[0096] P = klog(I / I0) ……(1) However, P: sensation intensity (also referred to as sensation amount), I: stimulus intensity, I0: stimulus intensity at which the sensation intensity becomes 0, k: constant specific to the stimulus. Details of these will be described later.

[0097] S5: Then, the mathematical model generation unit 65 constructs a mathematical model that outputs the exacerbation risk from the statistic of the environmental factor data and the sensation intensity, using the statistic of the environmental factor data and the sensation intensity as explanatory variables and the symptom change information as the objective variable (teacher data). Assuming that there are 6 types of environmental factor data, 4 statistics for each environmental factor data, and 1 sensation intensity for each environmental factor data, 6×5 = 30 data is one sample of the learning data.

[0098] In the present disclosure, although the statistic of the environmental factor data is used for learning, since the environmental factor data is also included in the sensation intensity, the mathematical model generation unit 65 may learn without using the statistic of the environmental factor data (using the sensation intensity).

[0099] The process of generating a mathematical model is also called machine learning. Machine learning is a technology for enabling a computer to acquire learning capabilities similar to those of humans. It refers to a technology in which a computer autonomously generates an algorithm necessary for judgments such as data identification from learning data pre-loaded and applies this to new data for prediction. The method using machine learning may be any of supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and deep learning. Furthermore, a learning method that combines these learning methods may also be used, and any learning method for machine learning is acceptable.

[0100] There are various algorithms for the method of generating a mathematical model using machine learning. In the present disclosure, for example, gradient boosting decision trees will be described. Gradient boosting decision trees are one of the supervised learning methods that combine "gradient descent method", "Boosting (ensemble)", and "decision trees".

[0101] FIG. 8 is a flowchart diagram for explaining the flow of the learning method using gradient boosting decision trees. FIG. 9 shows an image of a decision tree.

[0102] The mathematical model generation unit 65 calculates an initial value from a plurality of symptom change information (teacher data) prepared as learning data (S11). The initial value is, for example, an average value. For the sake of explanation, this average value is referred to as predicted value 1.

[0103] Next, the mathematical model generation unit 65 calculates, for each piece of learning data, a value obtained by subtracting the average value from the symptom change information as error 1 (S12). Error 1 is a positive value if the symptom change information is greater than the average and a negative value if it is smaller. Here, the reason for taking "the value obtained by subtracting the average value from the symptom change information" as the error is that the mean squared error between the symptom change information and the predicted value is used as the error calculation method, and the gradient obtained by differentiating the mean squared error is used for error calculation. The absolute value error may also be used as the error calculation method.

[0104] Next, the mathematical model generation unit 65 constructs a decision tree for the purpose of predicting errors (S13). The number of nodes and levels of the decision tree may be restricted to a preset range. This decision tree is a weak classifier (boosting). The top tree in FIG. 9 is an image of the initially created decision tree. Since FIG. 9 is an image, the number of nodes etc. is insufficient for the number of input data, but the errors are classified at the leaves (ends of the tree) of such a decision tree.

[0105] A brief explanation of the construction of the decision tree will be given. (i) The mathematical model generation unit 65 calculates the entropy based on the ratio of correct and incorrect answers of the learning data before classification. The mathematical model generation unit 65 may use the Gini coefficient instead of the entropy. (ii) The mathematical model generation unit 65 calculates the entropy for each branch based on the ratio of correct and incorrect answers of each branch when classified by any one attribute. (iii) The mathematical model generation unit 65 calculates the weighted average of the entropy in (ii). This weighted average may be the product of the ratio of the number of data classified into each branch to the original number of data. (iv) The mathematical model generation unit 65 adopts the attribute with the largest difference (gain) in entropy between (i) and (iii) as the root attribute. (v) The structure of the part under the root is also determined by the processes in (i) to (iv).

[0106] Next, the mathematical model generation unit 65 calculates a new predicted value using error 1 (S14). Here, it is referred to as "error 1", but "error n" increases by repetition. Since error 1 is classified for each sample of the learning data at the leaves of the decision tree, the predicted value 1 can be calculated for each sample of the learning data using error 1. When multiple errors are classified in one leaf, the average of the errors included in the leaf is error 1.

[0107] If the new predicted value is set as predicted value 2, then "predicted value 2 = predicted value 1 + learning rate * error 1". The learning rate is a value smaller than 1 and is a hyperparameter that determines how much the error is corrected in a single decision tree. The learning rate is appropriately determined, for example, to be 0.05 - 0.3. By repeatedly performing the operation of "calculating the error, multiplying it by the learning rate, and adding", the accuracy is gradually improved.

[0108] Next, the mathematical model generation unit 65 calculates the error from the symptom change information, which is the teacher data, and the predicted value 2 (S15). The mathematical model generation unit 65 calculates the error from the predicted value 2 for the symptom change information of all the prepared learning data. The error calculated from the predicted value 2 is called error 2.

[0109] The mathematical model generation unit 65 repeats steps S13 - S15 for a certain number of times or until the error becomes less than the threshold value (S16). As shown in the second and third trees in FIG. 9, the mathematical model generation unit 65 classifies the previous error with a new decision tree and calculates predicted values 3, 4... n and errors 3, 4... n using the error. Since the error also changes with the repetition of the process, the structure of the decision tree also automatically changes. The n decision trees created in this way are the mathematical models of the present disclosure. The prediction of the exacerbation risk using the n decision trees in the prediction phase will be described later.

[0110] As various frameworks for implementing the gradient boosting decision tree in an information processing device, LightGBM (Light Gradient Boosting Machine), XGBoost (eXtreme Gradient Boosting), Catboost (Category Boosting), etc. are known. The mathematical model generation unit 65 may use these frameworks.

[0111] Moreover, predicting the worsening risk using gradient boosting decision trees is a so-called regression problem (using continuous values to predict another numerical value from one or more numerical values), so the applicable algorithms are not limited to gradient boosting decision trees. There are many algorithms applicable to regression, such as linear regression (multiple regression), logistic regression, neural networks, Bayesian linear regression, SVM regression, ridge regression, lasso regression, Poisson regression, etc.

[0112] For example, deep learning is an algorithm that, after predicting XYZ based on the input data ABC, adjusts the weights between neural networks by the error backpropagation method to reduce the error from the teacher data.

[0113] <Physiological response amount to the environment based on the model simulating biological characteristics> The model that simulates biological characteristics and converts environmental factor data into intermediate data will be described in detail. Simulating biological characteristics means estimating the effects (such as physiological response amounts and behaviors) that humans receive from environmental factor data. Note that the Weber-Fechner law, which is one of these models, will be described in detail later.

[0114] Environmental factor data is converted into intermediate data by a model representing the following non-linear relationship. That is, the control unit 110 generates intermediate data correlated with the physiological response amount to the environment based on these models that simulate biological characteristics. The following models are all non-linear processing with quantitative information.

[0115] (i) A model that outputs intermediate data according to the ratio of the values taken by environmental factor data (logarithmic function, exponential function, n-th order function) This model is suitable for representing the relationship between the amount of dust and asthma symptoms, etc. The change according to the ratio of values means that when the environmental factor data increases at a certain ratio, the value of the intermediate data also increases at a predetermined ratio. Figures 10(a) to 10(c) show an example of the shapes of the logarithmic function 331, exponential function 332, and n-th order function 333, which are examples of this model.

[0116] (ii) A model that outputs different intermediate data values based on the range of values taken by the environmental factor data or whether the environmental factor data satisfies a condition (step function, sigmoid function, IF function, function with values only in a specific range) This model is suitable for representing relationships such as sweating and shivering due to temperature, or the relationship between temperature and the activity level of immune cells. Also, this model is suitable for representing relationships such as the relationship between temperature and humidity and heat stroke, and the relationship between the amount of dust and coughing and sneezing. Figures 11(a) to 11(d) show examples of the shapes of a step function 334, a sigmoid function 335, an IF function 336, and a function 337 with values only in a specific range, which are examples of this model. Note that the IF function 336 is a function that takes a constant value when the environmental factor data satisfies a predetermined condition and takes another constant value when it does not satisfy the predetermined condition.

[0117] (iii) A model that outputs intermediate data according to the way the environmental factor data changes rather than its absolute value (nth derivative (slope, acceleration), function with values only for one-way changes) This model is suitable for representing physiological reactions that are not noticed when the temperature change is small, or physiological reactions such as burns when touching a high temperature and not recovering even when the temperature returns. Also, this model is suitable for representing relationships such as the relationship between atmospheric pressure and meteorological diseases, or the relationship between acceleration and car sickness. Figure 12 shows an example of the shape (tangent line) of the first derivative 338 with respect to a quadratic function.

[0118] Figures 13(a) and 13(b) are diagrams for explaining a function with values only for one-way changes. Figure 13(a) shows the relationship between time and environmental factor data (for example, atmospheric pressure), and Figure 13(b) shows the relationship between time and intermediate data (for example, the amount of physiological reaction related to meteorological diseases). For example, when the environmental factor data is atmospheric pressure, the amount of physiological reaction changes only when the atmospheric pressure decreases. That is, intermediate data corresponding to the increase amount or decrease amount is output only when the environmental factor data changes either increase or decrease. Therefore, this model is suitable for representing relationships such as the relationship between atmospheric pressure and meteorological diseases.

[0119] (iv) A model that outputs intermediate data according to the integrated value of environmental factor data, or intermediate data that varies depending on the change history (hysteresis) of past values of environmental factor data even if the current values of the environmental factor data are the same. This model is suitable for representing relationships such as the amount of pollen intake and the amount of antibodies, or the relationship between CO2 concentration and blood oxygen concentration. Also, this model is suitable for representing relationships such as the relationship between the amount of pollen scattering and hay fever, or the relationship between temperature and humidity and heat stroke.

[0120] Figures 14(a) and (b) are diagrams for explaining intermediate data according to the integrated value of environmental factor data. Figure 14(a) shows the relationship between time and environmental factor data (for example, the amount of pollen scattering), and Figure 14(b) shows the relationship between time and intermediate data (for example, a physiological reaction related to hay fever). For example, if the environmental factor data is the amount of pollen scattering, the intermediate data increases according to the integrated value even if there is no change in pollen concentration (time T1), and does not change even if the pollen concentration becomes zero as long as the integrated value does not change (time T2). Therefore, this model is suitable for representing relationships such as the amount of pollen intake and the amount of antibodies, or the relationship between CO2 concentration and blood oxygen concentration.

[0121] Figures 15(a) and (b) are diagrams for explaining intermediate data that varies depending on the history (hysteresis) of past measurements of environmental factor data even if the environmental factor data is the same. Figure 15(a) shows the relationship between time and environmental factor data (for example, air temperature), and Figure 15(b) shows the relationship between time and intermediate data (for example, the amount of physiological reaction related to heat stroke). For example, if the environmental factor data is air temperature and the intermediate data is the amount of physiological reaction related to heat stroke. Even when the air temperature starts to drop, the amount of physiological reaction does not immediately drop (time t1), and even when the air temperature drops to the same temperature, the amount of physiological reaction does not drop to the same value (time t2). Therefore, this model is suitable for representing relationships such as the relationship between the amount of pollen scattering and hay fever, or the relationship between temperature and humidity and heat stroke.

[0122] (v) A model that outputs intermediate data according to the duration for which environmental factor data continues within a predetermined range of values, or the number of times environmental factor data repeats within a predetermined range of values. This model is suitable for representing the adaptation to temperature. Also, this model is suitable for representing the relationships such as the number of times of allergen intake and the amount of immune response (anaphylaxis).

[0123] Figures 16(a) and (b) are diagrams for explaining intermediate data according to the duration and the value that the environmental factor data is continuously measured. Figure 16(a) shows the relationship between time and environmental factor data (for example, CO2 concentration), and Figure 16(b) shows the relationship between time and intermediate data (for example, the amount of physiological reaction related to reduced wakefulness). For example, taking the environmental factor data as CO2 concentration and the intermediate data as the amount of physiological reaction related to reduced wakefulness. The amount of physiological reaction increases when the time with CO2 above a predetermined concentration lasts for a certain period or more (time T1), and decreases when the time with CO2 below the predetermined concentration lasts for a certain period or more (time T2). Therefore, this model is suitable for presenting the symptoms caused by the duration and the value to which the environmental factor data is continuously exposed.

[0124] Figures 17(a) and (b) are diagrams for explaining intermediate data according to the number of times the environmental factor data is measured. Figure 17(a) shows the relationship between time and environmental factor data (for example, the amount of allergen), and Figure 17(b) shows the relationship between time and intermediate data (the amount of physiological reaction related to allergic symptoms). For example, taking the environmental factor data as the amount of allergen and the intermediate data as the amount of physiological reaction related to allergic symptoms. The amount of physiological reaction is greater in the second exposure than in the first exposure even when the amount of allergen is the same. Therefore, this model is suitable for presenting the symptoms that occur according to the number of times of exposure to the environmental factor data.

[0125] (vi) A model that outputs intermediate data that changes after a predetermined time has elapsed since the time when the environmental factor data occurred This model is suitable for representing relationships such as a delay in nasal discharge after inhaling pollen. Also, this model is suitable for representing relationships such as a delay in the immune response after contact with bacteria.

[0126] Figures 18(a) and (b) are diagrams for explaining intermediate data corresponding to the measured environmental factor data after a lapse of time since the environmental factor data was measured. Fig. 18(a) shows the relationship between time and environmental factor data (for example, the amount of house dust), and Fig. 18(b) shows the relationship between time and intermediate data (for example, the amount of physiological reaction related to allergic symptoms). For example, the environmental factor data is the amount of house dust, and the intermediate data is the amount of physiological reaction related to allergic symptoms. The amount of physiological reaction is almost proportional to the amount of house dust, but changes with a slight delay (delay of time T) from the time when the amount of house dust changes. Therefore, this model is suitable for representing symptoms that occur after a lapse of time since being exposed to environmental factor data.

[0127] <Regarding the sensory intensity> Next, the Weber-Fechner law will be described in detail as one of the models simulating biological characteristics. According to the Weber-Fechner law, sensory intensity is obtained as an example of the above intermediate data. That is, the present disclosure introduces "sensory intensity" as a variable representing the magnitude of the influence that a human will receive when the change in the air quality environment serves as a stimulus. The sensory intensity is estimated from the air quality environment data based on the Weber-Fechner law proposed by Weber and Fechner.

[0128] The Weber-Fechner law is a law that expresses the intensity of a stimulus felt by a human in a mathematical formula and is said to approximately apply to all five senses. The sensory intensity P is estimated from the statistic of the environmental factor data as shown in Equation (1). I, I0, and k in Equation (1), which are necessary for estimating the sensory intensity from the air quality environment data, are determined as follows. The "intensity of the stimulus I" uses the statistic of the environmental factor data. The "intensity of the stimulus at which the intensity of the sensation becomes 0 (the intensity of the stimulus at which the stimulus begins to be felt) I0" uses the average value of the environmental factor data. The reason for using the average value is as follows.

[0129] It is known that humans have the characteristic of adapting to the surrounding environment. This adaptation progresses towards the average state of the surrounding environment. Considering these factors, in the present invention, the average value is used as the intensity I0 of the stimulus at which the stimulus begins to be felt. Regarding the "constant k specific to the stimulus", although it is set to different values for each sense, in the present invention, it is used as a weighting coefficient to equalize the differences in the degrees of influence caused by the different ranges and units that the environmental factor data (for example, temperature, humidity, CO2, PM2.5, TVOC, formaldehyde, etc.) can take, and is determined for each environmental factor. The determination method is to create an approximate formula with the environmental factor data as the explanatory variable and the exacerbation risk as the objective variable, and obtain it as a constant that minimizes the approximation error.

[0130] Figures 19(a) and (b) are diagrams showing the correspondence between environmental factor data (X-axis) and sensory intensity (Y-axis). Figure 19(a) is a graph of the following formula (2). Y = k log X + α......(2) However, k and α may be appropriately designed constants.

[0131] Note that formulas (1) and (2) are examples, and the correspondence between the stimulus and the sensory intensity may be represented by, for example, formula (3). Y = log I......(3) I is the statistic of the environmental factor data / the average value per day.

[0132] Also, the sensory intensity may be obtained other than by logarithm. Figure 19(b) shows the correspondence between the stimulus and the sensory intensity by two straight lines. In this way, the sensory intensity may saturate in the region where the stimulus is larger than in the region where the stimulus is smaller. The correspondence between the stimulus and the sensory intensity may be represented by three or more straight lines.

[0133] Figure 20 is a diagram explaining that the sensory intensity is estimated from the environmental factor data and corresponds to the biological reaction. In Figure 20, the surrounding environment A and the human side (inside the ecosystem) B are shown separately.

[0134] Surrounding environment A: Humans feel stimuli due to environmental changes such as temperature changes, humidity changes, air quality changes, and odor changes.

[0135] Human side (inside the ecosystem) B: According to Weber-Fechner's law, the stimulus increases and saturates. That is, humans perceive the intensity of sensation rather than the magnitude of the stimulus. And the intensity of sensation appears as various biological reactions (changes in heart rate, blood pressure, respiratory rate, etc.) that change the physical form. Therefore, it can be seen that the intensity of sensation is information that can replace vital data.

[0136] <Prediction of exacerbation risk> Subsequently, a prediction method for predicting the exacerbation risk from environmental factor data and the intensity of sensation using the mathematical model generated by the mathematical model generation unit 65 will be described.

[0137] <<Regarding functions>> FIG. 21 is an example of a functional block diagram for explaining the functions of the information processing apparatus 60 by dividing them into blocks in the prediction phase. In the description of FIG. 21, the differences from FIG. 5 will be mainly described. The information processing apparatus 60 includes an environmental factor data acquisition unit 61, a statistic calculation unit 63, a sensation intensity estimation unit 64, and an exacerbation risk prediction unit 66. Each of these units included in the information processing apparatus 60 is a function or means realized by the control unit 110 of the information processing apparatus 60 executing the instructions of the program developed in the memory 222.

[0138] Among these, the deterioration risk prediction unit 66 corresponds to a mathematical model. The deterioration risk prediction unit 66 predicts the possibility of symptom change from the intermediate data by using the correspondence information associating the intermediate data with the possibility of symptom change. In the present disclosure, the deterioration risk prediction unit 66 outputs a deterioration risk from environmental factor data (actual measurement, forecast) and the perceived intensity estimated based on the environmental factor data. The possibility of symptom change indicates the degree of probability that the symptom will deteriorate or improve. Taking the deterioration risk as an example, when the symptom change information input by the user is 1 (deteriorated) or 0 (not deteriorated), the predicted value of the possibility of symptom change also takes a value in the range of 0 to 1. The closer the predicted value is to 1, the more likely the symptom will change to a worse state, and the closer the predicted value is to 0, the more likely the symptom will change to a better state. The deterioration risk unit 66 can predict how much the symptom will change. When using the forecast value for the environmental factor data, the perceived intensity is also estimated using the forecast value. The deterioration risk is the value predicted by the deterioration risk prediction unit 66 for the self-reported symptom change information.

[0139] FIG. 22 is a diagram for explaining a prediction phase in which environmental factor data and the perceived intensity are input to the deterioration risk prediction unit 66 to predict the deterioration risk. In the description of FIG. 22, the differences from FIG. 7 will be mainly described. Steps S1, S3, and S4 may be the same as those in FIG. 7.

[0140] S6: The deterioration risk prediction unit 66 outputs a deterioration risk using the statistic of the environmental factor data and the perceived intensity estimated from the statistic as input data.

[0141] The prediction using the gradient boosting decision tree will be described. The deterioration risk prediction unit 66 inputs the input data to all the decision trees created in the learning phase. In FIG. 9, there are three decision trees, and for each decision tree, the input data is classified into one leaf of the decision tree. An error is stored in each leaf. For example, it is assumed that errors 1 to 3 are classified into the leaves 321 to 323 indicated by the dotted circles in FIG. 9. The deterioration risk prediction unit 66 estimates the predicted value (deterioration risk) of the symptom change information by summing up, for all the decision trees, the value obtained by multiplying the learning rate by the errors of these leaves with respect to the average value calculated in the learning phase.

[0142] Predicted value of deterioration risk = average + learning rate × error 1 + learning rate × error 2 + learning rate × error 3 In this way, in the gradient boosting decision tree, the final predicted value is the value obtained by adding all of error 1 to error n, each multiplied by the learning rate, to the average calculated in the learning phase. Final predicted value = average + learning rate × error 1 + learning rate × error 2 + …… + learning rate × error n When the symptom change information in the training data has deteriorated (1) or has not deteriorated (0), the predicted value is a value between 0 and 1. The deterioration risk prediction unit 66 may multiply the predicted value by 100 to convert it to a percentage display.

[0143] <Improvement in prediction accuracy by using sensory intensity> With reference to FIG. 23, the effect of using sensory intensity will be described. FIG. 23 is a diagram comparing the prediction accuracy when predicting the deterioration risk of asthma without using sensory intensity and when using it. The prediction accuracy when predicting without using sensory intensity is 61.1%, whereas the prediction accuracy when predicting using sensory intensity has improved to 73.3%. <Modification example of the learning phase> It is known that symptom change information has large individual differences. This means that in the learning phase, it is effective for the mathematical model generation unit 65 to generate a mathematical model for each individual. The mathematical model generation unit 65 generates a mathematical model (an example of the first correspondence information) for all patients who report symptom change information, and also generates a mathematical model (an example of the second correspondence information) for each individual.

[0144] The deterioration risk prediction unit 66 predicts the deterioration risk using the mathematical models of multiple people (for example, all patients), and also predicts the deterioration risk using the individual's mathematical model. The deterioration risk prediction unit 66 provides one or more of the two predicted deterioration risks, the higher of the two predicted deterioration risks, or the average of the two predicted deterioration risks to the individual, making it easier for each individual to grasp the deterioration risk for the day.

[0145] In addition, since some allergens, such as pollen, become environmental factors depending on the season, it is effective for the mathematical model generation unit 65 to generate a mathematical model for each season. Also, since allergens vary from person to person, it is effective for the mathematical model generation unit 65 to generate a mathematical model for each allergen. The allergens of patient 9 are registered in the information processing device 60. The mathematical model generation unit 65 groups patients 9 with the same allergen and generates a mathematical model based on the reported symptom change information and environmental factor data. Also, since the diseases (underlying diseases) that are likely to develop vary from person to person, it is effective for the mathematical model generation unit 65 to generate a mathematical model for each target disease. The diseases of patient 9 are registered in the information processing device 60. The mathematical model generation unit 65 groups patients 9 with the same disease and generates a mathematical model based on the reported symptom change information and environmental factor data.

[0146] When sufficient environmental factor data is obtained, further, the creation of the mathematical model may be performed for each user attribute (for example, gender and age) and area (for example, prefecture and municipality). Thereby, an improvement in the prediction accuracy of the exacerbation risk can be expected.

[0147] <Example of presenting exacerbation risk> FIG. 24 is a display example of a map screen 300 of the exacerbation risk displayed on the user terminal 70. The map screen 300 mainly has a mode selection bar 301 and a map display bar 302. The mode selection bar 301 has a recommendation button 303, a mold & mite button 304, an energy saving button 305, and a decision button 306.

[0148] · The recommendation button 303 is a button that displays a recommended environmental setting by comprehensively considering mold & mite resistance and energy saving.

[0149] · The mold & mite button 304 is a button that displays an environmental setting in which mold and mites are suppressed based on at least one of the mold index or the mite index.

[0150] · The energy-saving button 305 is a button for displaying environmental settings recommended based on energy-saving performance.

[0151] · The decision button 306 is a button for accepting the control of environmental equipment with the environmental settings set by the patient 9.

[0152] The map display column 302 displays three maps A to C. This is an example, and maps A to C may be displayed one by one. In the three maps A to C, the current environmental situations 308 to 310 are shown respectively.

[0153] · Map A shows an area with a high exacerbation risk associated with the set temperature and set humidity. The area with a low exacerbation risk is divided into two parts. Corresponding to the temperature and humidity of the map screen original data, the area where the exacerbation risk is below the first threshold is shown in blue (an example), and the area where the exacerbation risk is below the second threshold is shown in red (an example). Note that the map screen original data is data in which environmental factor data and exacerbation risks are comprehensively calculated. Since the map screen original data is discrete data, the screen generation unit 73 performs processes such as interpolation between points and coloring.

[0154] · Map B shows an area with a low exacerbation risk associated with the CO2 concentration and PM2.5 concentration. The area with a low exacerbation risk is divided into two parts. Corresponding to the CO2 concentration and PM2.5 concentration of the map screen original data, the area where the exacerbation risk is below the first threshold is shown in blue (an example), and the area where the exacerbation risk is below the second threshold is shown in red (an example). However, the first threshold < the second threshold.

[0155] · Map C shows an area with a low exacerbation risk associated with the formaldehyde concentration and TVOC concentration. The area with a low exacerbation risk is divided into two parts. Corresponding to the formaldehyde concentration and TVOC concentration of the map screen original data, the area where the exacerbation risk is below the first threshold is shown in blue (an example), and the area where the exacerbation risk is below the second threshold is shown in red (an example).

[0156] Therefore, the user can easily determine what temperature and humidity the air conditioner should be set to.

[0157] Note that instead of the user terminal 70 displaying the map screen, the information processing device 60 may display the map screen on the display device 230.

[0158] FIG. 25 is an example of the display of the perceived intensity and the exacerbation risk screen 310 displayed by the user terminal 70. The perceived intensity and the exacerbation risk screen 310 display a message 311 saying "The exacerbation risk of allergic symptoms is as follows", an exacerbation risk 312, and the degree of influence of environmental factors involved in the exacerbation risk 313.

[0159] The user can check the displayed exacerbation risk 312 and visually grasp the environmental factors involved in the exacerbation risk. Further, the perceived intensity and the exacerbation risk screen 310 display environmental factors 314 whose degree of influence involved in the exacerbation risk is equal to or higher than the threshold value. Thereby, the user can know which environmental factors need to be adjusted.

[0160] <More detailed description of environmental factor data and symptoms> FIGS. 26(a) and (b) show an example of environmental factor data and symptoms. FIG. 26(a) shows environmental factor data detectable by an environmental sensor. As environmental factor data, in addition to temperature, humidity, pollen, mold, mites, there are also radiant temperature, atmospheric pressure, dust (amount, floating concentration), air flow (wind speed, air volume), CO2, oxygen (concentration), sound (sound pressure, volume, frequency, tempo), odor (intensity), acceleration and inclination (environmental factor data when riding in a vehicle), droplets, vapor, smoke (environmental factor data related to infectious diseases), etc. In the present disclosure, these can be the original data for predicting the possibility of symptom changes.

[0161] FIG. 26(b) shows the symptoms caused by the environmental factor data. As symptoms, in addition to allergies, there are also meteorological diseases, infectious diseases, sleep quality, autonomic nervous disorders, dementia and delirium, frailty, decline and improvement of memory, arousal and drowsiness, heat stroke, motor function, car sickness, and VR sickness, etc. In the present disclosure, these can also be symptom changes predicted by a mathematical model. When the mathematical model is being trained, the user reports the presence or absence and degree of these symptoms as teacher data.

[0162] <Main effects> As described above, in the present disclosure, vital data is not used in predicting the exacerbation risk. Further, the present disclosure can replace vital data and accurately predict the exacerbation risk of symptoms by introducing "sensation intensity" as a variable representing the magnitude of the influence that a human will receive, with the change in the air quality environment as a stimulus. Also, by converting environmental factor data into intermediate data using a model that simulates biological characteristics, the accuracy of machine learning can be improved with a small number of samples.

[0163] <Other application examples> As described above, the best mode for carrying out the present disclosure has been described using examples. However, the present disclosure is not limited to such examples, and various modifications and substitutions can be made without departing from the gist of the present disclosure.

[0164] For example, the information processing device 60 and the environmental sensor 11 may not be a separate system, and the information processing device 60 may be integrated with the environmental sensor 11. In this case, the environmental sensor 11 can display the exacerbation risk predicted from the environmental factor data of the installed space on a liquid crystal display or the like.

[0165] Similarly, when the environmental sensor 11 is built into the indoor unit 10b of the air conditioner, any one of the indoor unit 10b, the outdoor unit 10a, or the remote control 12 may display the exacerbation risk predicted from the environmental factor data of the space on the remote control 12.

[0166] Also, in the present disclosure, vital data is not required for predicting the exacerbation risk, but the information processing device 60 may predict the exacerbation risk using the sensation intensity and vital data.

[0167] Also, this embodiment is not limited to predicting human symptoms, and may be used to predict symptoms of animals such as pets, livestock such as cows and pigs, and farmed fish.

[0168] Also, the configuration examples in FIGS. 5, 21, etc. are divided according to the main functions in order to facilitate the understanding of the processing by the information processing apparatus 60. The technology of the present disclosure is not limited by the way of dividing the processing units and the names. The processing of the information processing apparatus 60 can be further divided into more processing units according to the processing content. Also, one processing unit can be divided so as to include more processing.

[0169] Also, the device group described in the embodiments only shows one of a plurality of computing environments for implementing the embodiments disclosed in this specification. In one embodiment, the information processing apparatus 60 includes a plurality of computing devices such as a server cluster. The plurality of computing devices are configured to communicate with each other via an arbitrary type of communication link including a network, a shared memory, etc., and implement the processing disclosed in this specification.

[0170] Each function of the present disclosure described above can be realized not only by software processing by executing a program but also by one or a plurality of processing circuits. Here, the "processing circuit" in this specification includes a processor programmed to execute each function by software like a processor implemented by an electronic circuit, an ASIC (Application Specific Integrated Circuit) designed to execute each function described above, a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), and devices such as conventional circuit modules.

[0171] <Reason for the effect> ·The first aspect of the present disclosure has a control unit that converts the environmental factor data into intermediate data by a model simulating biological characteristics related to symptoms for the environment, Since the control unit predicts the possibility of symptom change from the intermediate data using correspondence information that associates at least the intermediate data with the possibility of symptom change in the environment, even if there is no simple linear relationship between the symptoms that occur in humans and environmental factors and there is a complex correspondence relationship, correspondence information associating the intermediate data with the possibility of symptom change in the environment can be created with a smaller number of samples of learning data.

[0172] · Since the model simulating the biological characteristics in the second aspect of the present disclosure is Weber-Fechner's law, the change in the air quality environment can be appropriately converted into intermediate data as a stimulus and the magnitude of the influence that humans will receive, and correspondence information associating the intermediate data with the possibility of symptom change in the environment can be created with a smaller number of samples of learning data.

[0173] · Since the model simulating the biological characteristics in the third aspect of the present disclosure is a model that outputs different intermediate data depending on the range of values taken by the environmental factor data or whether the environmental factor data satisfies a condition, sweating or trembling due to temperature, or the relationship between temperature and the activity level of immune cells, etc. can be represented by intermediate data, and correspondence information associating the intermediate data with the possibility of symptom change in the environment can be created with a smaller number of samples of learning data.

[0174] · Since the model simulating the biological characteristics in the fourth aspect of the present disclosure is a model that outputs intermediate data corresponding to an increase amount or a decrease amount only when the change in the environmental factor data changes either to an increase or a decrease, or a model that outputs intermediate data that changes according to the value of the nth derivative, physiological reactions that are not noticed when the temperature change is small, physiological reactions such as burns that do not recover even when the temperature returns after touching a high temperature, the relationship between atmospheric pressure and altitude sickness, or the relationship between acceleration and car sickness can be represented by intermediate data, and correspondence information associating the intermediate data with the possibility of symptom change in the environment can be created with a smaller number of samples of learning data.

[0175] · Since the fifth aspect of the present disclosure is "a model that simulates biological characteristics outputs the intermediate data according to the integrated value of the environmental factor data, or the intermediate data that is different depending on the change history of the past values of the environmental factor data even if the current value of the environmental factor data is the same", relationships such as the relationship between pollen intake and antibody amount, CO2 concentration and blood oxygen concentration, the relationship between pollen dispersal amount and hay fever, or the relationship between temperature and humidity and heat stroke can be represented by intermediate data, and correspondence information associating the intermediate data with the possibility of symptom changes in the environment can be created with a smaller number of samples of learning data.

[0176] · Since the sixth aspect of the present disclosure is "a model that simulates biological characteristics outputs the intermediate data according to the duration for which the environmental factor data continues to be within a predetermined range of values, or the number of times the environmental factor data repeats within a predetermined range of values", relationships such as getting used to the temperature, the relationship between the number of allergen intakes and the amount of immune response (anaphylaxis) can be represented by intermediate data, and correspondence information associating the intermediate data with the possibility of symptom changes in the environment can be created with a smaller number of samples of learning data.

[0177] · Since the seventh aspect of the present disclosure is "a model that simulates biological characteristics outputs the intermediate data that changes after a predetermined time has elapsed since the time when the environmental factor data occurred", relationships such as the relationship where a runny nose appears after inhaling pollen with a delay, or the relationship where the immune system works with a delay after contact with bacteria can be represented by intermediate data, and correspondence information associating the intermediate data with the possibility of symptom changes in the environment can be created with a smaller number of samples of learning data.

[0178] · Since the eighth aspect of the present disclosure is "a model that simulates biological characteristics is a log function, exponential function, or nth-order function that takes the environmental factor data as input and outputs the intermediate data", relationships such as the relationship between the amount of dust and asthma symptoms can be represented by intermediate data, and correspondence information associating the intermediate data with the possibility of symptom changes in the environment can be created with a smaller number of samples of learning data.

[0179] · Since the ninth aspect of the present disclosure is that "the control unit further predicts the possibility of symptom change from the intermediate data and the environmental factor data using the correspondence information associated with the environmental factor data", it is possible to predict the possibility of symptom change not only from the intermediate data but also from the environmental factor data.

[0180] · Since the tenth aspect of the present disclosure is that "the possibility of symptom change is the possibility that any of the symptoms of allergic symptoms, asthma symptoms, meteoropathy, infectious diseases, sleep quality deterioration, awakening deterioration / sleepiness, autonomic nerve disorder, frailty, dementia / delirium, memory deterioration, motor function deterioration, heat stroke, motion sickness, or VR sickness deteriorates or improves", it is possible to predict the possibility of change in various symptoms.

[0181] · Since the eleventh aspect of the present disclosure is that "the environmental factor data is a statistic processed by statistical processing", it is possible to convert the statistical processing into intermediate data instead of the environmental factor data itself, and since the correspondence information associates this intermediate data with the possibility of symptom change in the environment, the accuracy of prediction can be improved.

[0182] · Since the twelfth aspect of the present disclosure is that "the possibility of symptom change is predicted from the actually measured environmental factor data and the intermediate data, or from the forecast value of the environmental factor data and the intermediate data", it is possible to predict the possibility of symptom change not only from the actually measured environmental factor data and the intermediate data but also from the forecast value of the environmental factor data and the intermediate data.

[0183] · Since the thirteenth aspect of the present disclosure is that "using the intermediate data as an explanatory variable and the symptom change information reported for asthma symptoms, allergic symptoms, meteoropathy, infectious diseases, sleep quality deterioration, awakening deterioration / sleepiness, autonomic nerve disorder, frailty, dementia / delirium, memory deterioration, motor function deterioration, heat stroke, motion sickness, or VR sickness as teacher data, the correspondence information is generated using a machine learning method", it is possible to generate correspondence information that has learned the correspondence between the explanatory variable and the teacher data, and the possibility of these symptom changes can be predicted with this correspondence information.

[0184] · The 14th aspect of the present disclosure is to "predict the possibility of an individual's symptom change based on the possibility of symptom change predicted by the first correspondence information for multiple persons and the second correspondence information for an individual", so correspondence information for multiple persons and for an individual can be generated respectively, and the possibility of symptom change is predicted with each of the two pieces of correspondence information, so it is possible to provide, for each individual, the one with a higher risk of exacerbation, etc.

[0185] · The 15th aspect of the present disclosure is to "predict the possibility of symptom change based on the correspondence information generated for each season, for each target disease, or for each of the allergens", so it is possible to generate for each season, for each target disease, or for each allergen and predict the possibility of symptom change.

[0186] · The 16th aspect of the present disclosure is to "display the intermediate data and the possibility of symptom change on the same screen", so it is easy to grasp how likely there is a symptom change with respect to the current intermediate data.

Explanation of Signs

[0187] 10 Environmental device 11 Environmental sensor 60 Information processing device 70 User terminal 100 Symptom change prediction system for persons

Claims

1. A biological symptom change prediction system having an environmental sensor and an information processing device, wherein the environmental sensor detects environmental factor data regarding environmental factors in a target space, the information processing device has a control unit that converts the environmental factor data into intermediate data by a model simulating biological characteristics related to symptoms with respect to the environment, the control unit predicts the possibility of symptom change from the intermediate data using correspondence information associating at least the intermediate data with the possibility of symptom change with respect to the environment. A biological symptom change prediction system.

2. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics is Weber-Fechner's law.

3. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics outputs different intermediate data depending on the range of values taken by the environmental factor data or whether the environmental factor data satisfies a condition.

4. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics outputs the intermediate data corresponding to the increase amount or decrease amount only when the change in the environmental factor data changes either to increase or decrease, or outputs the intermediate data that changes according to the value of the nth derivative.

5. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics outputs the intermediate data corresponding to the integrated value of the environmental factor data, or outputs the intermediate data that is different depending on the change history of the past values of the environmental factor data even if the current value of the environmental factor data is the same.

6. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics outputs the intermediate data corresponding to the duration for which the environmental factor data continues within a predetermined range of values, or the number of times the environmental factor data repeats within a predetermined range of values.

7. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics outputs the intermediate data that changes after a predetermined time has elapsed since the time when the environmental factor data occurred.

8. The biological symptom change prediction system according to claim 1, wherein the model simulating the biological characteristics is a log function, an exponential function, or an nth-order function that takes the environmental factor data as an input and outputs the intermediate data.

9. The control unit further predicts the possibility of symptom change from the intermediate data and the environmental factor data by using correspondence information associated with the environmental factor data, according to any one of claims 1 to 8 of the biological symptom change prediction system.

10. The possibility of symptom change is the possibility that any one of allergic symptoms, asthma symptoms, meteoropathy, infectious diseases, sleep quality decline, wakefulness decline / sleepiness, autonomic nerve disorder, frailty, dementia / delirium, memory decline, motor function decline, heat stroke, motion sickness, or VR sickness worsens or improves, according to claim 1 of the biological symptom change prediction system.

11. The environmental factor data is a statistic processed by statistical processing, according to claim 1 of the biological symptom change prediction system.

12. The control unit predicts the possibility of symptom change from the actually measured environmental factor data and the intermediate data, or from the forecast value of the environmental factor data and the intermediate data, according to claim 9 of the biological symptom change prediction system.

13. The control unit uses the intermediate data as an explanatory variable, Regarding allergic symptoms, asthma symptoms, meteoropathy, infectious diseases, sleep quality decline, wakefulness decline / sleepiness, autonomic nerve disorder, frailty, dementia / delirium, memory decline, motor function decline, heat stroke, motion sickness, or VR sickness, the reported symptom change information is used as teacher data, and generates the correspondence information by using a machine learning method, according to claim 1 of the biological symptom change prediction system.

14. The control unit uses the intermediate data as an explanatory variable and the symptom change information reported by multiple people as teacher data, and generates first correspondence information by using a machine learning method, uses the intermediate data as an explanatory variable and the symptom change information reported by an individual as teacher data, and generates second correspondence information by using a machine learning method, and the control unit predicts the possibility of an individual's symptom change based on the possibility of symptom change predicted by the first correspondence information and the second correspondence information respectively, according to claim 13 of the biological symptom change prediction system.

15. The control unit generates the correspondence information for each season, for each target disease, or for each allergen in allergic symptoms, and predicts the possibility of symptom change based on the correspondence information generated for each season, for each target disease, or for each allergen, according to claim 13 or 14 of the biological symptom change prediction system.

16. The biological symptom change prediction system according to any one of claims 1 to 8, wherein the control unit displays the intermediate data and the possibility of symptom change on the same screen.

17. An information processing apparatus, receiving environmental factor data regarding environmental factors in a target space from an environmental sensor, having a control unit that converts the environmental factor data into intermediate data by a model simulating biological characteristics related to symptoms for the environment, wherein the control unit predicts the possibility of symptom change from the intermediate data using correspondence information associating at least the intermediate data and the possibility of symptom change for the environment.

18. A symptom change prediction method performed by a biological symptom change prediction system having an environmental sensor and an information processing apparatus, wherein the environmental sensor detects environmental factor data regarding environmental factors in a target space, the information processing apparatus has a control unit, the control unit performs a process of converting the environmental factor data into intermediate data by a model simulating biological characteristics related to symptoms for the environment, and a process of predicting the possibility of symptom change from the intermediate data using correspondence information associating at least the intermediate data and the possibility of symptom change for the environment, and is a symptom change prediction method for performing the above.

Images