WO2021140731A1 - Information transmitting device and information transmitting method - Google Patents

Information transmitting device and information transmitting method Download PDF

Info

Publication number
WO2021140731A1
WO2021140731A1 PCT/JP2020/041125 JP2020041125W WO2021140731A1 WO 2021140731 A1 WO2021140731 A1 WO 2021140731A1 JP 2020041125 W JP2020041125 W JP 2020041125W WO 2021140731 A1 WO2021140731 A1 WO 2021140731A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
data
specific
inference
inspection
Prior art date
Application number
PCT/JP2020/041125
Other languages
French (fr)
Japanese (ja)
Inventor
野中　修
智子 後町
弘達 藤原
亮 櫻井
Original Assignee
オリンパス株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by オリンパス株式会社 filed Critical オリンパス株式会社
Priority to CN202080085581.7A priority Critical patent/CN114868203A/en
Publication of WO2021140731A1 publication Critical patent/WO2021140731A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment

Definitions

  • the present invention relates to an information transmission device and an information transmission method capable of providing a user with customized information such as advice according to an inspection result that can be obtained in daily life.
  • the Internet has become widespread in recent years, and by using the Internet, it is possible to easily obtain information closely related to the user's life.
  • various customized information valid information
  • services that provide this customized information are increasing. For example, services that introduce health foods are often interesting information that is common to many people, so this type of service is often found.
  • Patent Document 1 discloses a remote inspection method for transmitting inspection data by using a sensor chip and a mobile phone as a reader / writer and using a public communication network. Then, in Patent Document 1, it is proposed to store past inspection data and its evaluation results in a database and use these information.
  • Patent Document 2 discloses a biometric information measuring device that integrates personal authentication data and image data photographed by excrement photographing means and transmits the integrated data by communication means. Further, in Patent Document 3, a study list related to an examination is displayed, a medical image based on the medical image information of the subject in the selected study is displayed on the image display screen, and when a history browser is requested, the study of the subject is requested. A method of displaying medical information that displays a history browser representing a list of medical information is disclosed.
  • Patent Document 4 discloses that information according to the patient's situation and symptoms is presented based on the patient's interview information, past diagnostic information such as test values, hospital information, and the like.
  • Patent Document 5 discloses that a medical person checks a change pattern of a patient's time-series examination data on a predetermined time axis.
  • Patent Document 6, Patent Document 7, and Patent Document 8 disclose that time series data is input to an inference engine to obtain an inference result.
  • Patent Document 9 discloses that it assists in determining the optimum inspection interval.
  • Patent Document 10 discloses that model learning is performed using the result data of biochemical tests of samples such as blood, urine, and stool, and information is provided using this model.
  • JP-A-2009-258886 Japanese Unexamined Patent Publication No. 2014-031655 Japanese Patent No. 5294947 Japanese Unexamined Patent Publication No. 2016-018457 Japanese Unexamined Patent Publication No. 11-089822 JP-A-2018-518207 International Publication No. 2019/022085 Japanese Unexamined Patent Publication No. 2006-511881 Japanese Unexamined Patent Publication No. 2011-253464 Japanese Unexamined Patent Publication No. 2019-21186
  • Patent Document 1-3 describes that biometric information is acquired and this information is remotely transmitted and used. If the subject has no subjective symptoms, it is useful to know that he / she is ill and that he / she may become ill. However, Patent Document 1-3 does not describe anything about generating customized information such as advice according to an individual's profile and notifying the user of this customized information.
  • Patent Document 6-8 discloses that inference is performed using an inference model and information is provided to the subject.
  • the inspection data may change significantly when there is a certain event (trigger).
  • a certain event Trigger
  • the impact event that affected the change in the user's health condition is known, the user can subsequently pay attention to the impact event and maintain his / her health.
  • the present invention has been made in view of such circumstances, and an object of the present invention is to provide an information transmission device and an information transmission method capable of presenting an event related to a change in a user's health condition.
  • the information transmission method acquires inspection data in a time series of a specific period using an inspection device having a specific specification, acquires change pattern information of the inspection data, and obtains the change pattern information of the inspection data. Inference that the effect event information at the time corresponding to the time when the change pattern changes within the specific period is detected for the change pattern information, and the result of annotating the detected effect event information is learned as teacher data.
  • An impact event is inferred by acquiring a model and inputting inspection data obtained in time series in a period similar to the width of the specific period using the inspection equipment of the specific specifications of a specific person into the inference model. And convey the result of the inference.
  • the information transmission method corresponds to the input of a specific health response event, and prior to the above health response event, inspection data is acquired and recorded in time series using an inspection device having specific specifications.
  • an inference model is acquired in which the result of annotating the equipment and / or equipment and / or environment information at the time of the above-mentioned health response event is trained as teacher data with respect to the change pattern information of the test data. Then, the inspection data obtained in time series using the inspection equipment of the specific specifications of the specific person is input to the inference model to infer the information of the equipment and / or the equipment and / or the environment. From the inference result, the health response event customized for the specific person is transmitted to the specific person.
  • the information transmission method includes change pattern information of the above data created by acquiring a plurality of inspection data in a time series over a specific period using an inspection device having a specific specification, and the above-mentioned identification. Create a database that can record the timing of each of the plurality of event information related to the acquisition source of the inspection data within the period of the above, and use the database to determine the trend change of the change pattern and the timing of the event. Based on the relationship, it is possible to extract the events that have influenced the above-mentioned trend change from the above-mentioned event information and provide the information.
  • the information transmission method according to the fourth invention is based on the above-mentioned component in the above-mentioned third invention, in which the above-mentioned event information is decomposed for each component and components are extracted, and the event information affecting the above-mentioned trend change is extracted. Customize and provide information.
  • the information transmission method is the inference model for providing an event influencing the trend change by using the database in the fourth invention.
  • the inference model acquires inspection data and obtains the inspection data.
  • a change pattern is used as an input of the inference unit for learning, and advice to be output is used as annotation information to generate an inference model.
  • the change pattern information of the target person is used as the inference model.
  • the inference result is obtained by inputting, and the above-mentioned transmission information is determined based on the obtained inference result.
  • the information transmission device includes a data acquisition unit that acquires inspection data in a time-series manner for a specific period using an inspection device having a specific specification, and acquires change pattern information of the inspection data, and the above-mentioned change.
  • a data acquisition unit acquires inspection data in a time-series manner for a specific period using an inspection device having a specific specification, and acquires change pattern information of the inspection data, and the above-mentioned change.
  • an inference model is obtained in which the influence event information at the time corresponding to the time when the change pattern changes within the specific period is detected and the result of annotating the detected influence event information is trained as teacher data.
  • the influence event is input by inputting the inspection data obtained in time series in the width of the specific period and the period similar to the width of the specific period using the learning unit to be acquired and the inspection device of the specific specification of the specific person into the inference model. It has an inference unit for inferring and an information transmission unit for transmitting the result of the inference.
  • the information transmission device corresponds to the input of a specific health response event, and data for acquiring and recording inspection data in a time series using an inspection device having specific specifications prior to the health response event.
  • the result of annotating the information on the equipment, / or equipment, and / or environment at the time of the above-mentioned health response event with respect to the change pattern information about the above-mentioned inspection data recorded by the acquisition unit is used as the teacher data.
  • Equipment and / or equipment by inputting the inspection data obtained in time series using the learning unit that acquires the trained inference model and the inspection equipment of the above-mentioned specific specifications of a specific person into the inference model. It also has an inference unit that infers environmental information and / or an information transmission unit that transmits a health response event customized for the specific person from the inference result by the inference unit to the specific person.
  • the information transmission device is created by a data acquisition unit that acquires a plurality of inspection data in a time-series manner over a specific period using an inspection device having a specific specification, and an inspection data that is acquired and created.
  • a data creation unit that creates a database that can record the change pattern information of the inspection data and the timing of each of a plurality of event information related to the acquisition source of the inspection data within the specific period, and the database. Based on the relationship between the trend change of the change pattern and the timing of the event, the information providing unit that can extract the event that affects the trend change from the event information and provide the information. Have.
  • an information transmission device and an information transmission method capable of presenting an event related to a change in a user's health condition.
  • the present invention as an example of grasping the accurate health condition by considering the situation of the subject and providing customized information, the profile information of the subject is stored, and the inspection data regarding the health condition is collected every day.
  • the subject in this embodiment is a person who may become a patient, depending on the re-examination.
  • this subject is also a person who can regain confidence in health, do not have to worry about illness, and enjoy daily life.
  • he is also a person who can become healthy by simple improvement and treatment of life.
  • the information transmission system has an inspection data acquisition unit (for example, refer to the information determination device 2 in FIG. 1) that acquires the inspection data of the subject, profile information of the subject, and information on the possessed equipment for each medical institution.
  • a storage unit for example, see DB 8 in FIG. 1) to be stored, and a transmission information determination unit (for example, a transmission information determination unit) that determines transmission information to the target person according to test data, profile information of the target person, and possessed device information for each medical institution.
  • This information transmission system is composed of, for example, a server, but may be configured by a personal computer capable of exchanging information with the server, a mobile information device such as a smartphone, or the like.
  • the information transmission system is sent to the inspection data acquisition unit (for example, refer to the information determination device 2 in FIG. 1) that acquires the inspection data of the subject, the time-series change pattern of the inspection data, and the medical institution of the subject. It has a transmission information determination unit (see, for example, control unit 1 and inference engine 7 in FIG. 1) that determines transmission information to a subject according to an inference model generated by machine learning using the visit information of the above.
  • the visit information is not limited to medical institutions, but also includes information when a person is examined at a medical institution or when a drug is purchased and taken at a pharmacy or the like.
  • This information transmission system is also composed of, for example, a server, but may also be composed of a personal computer capable of exchanging information with the server, a mobile information device such as a smartphone, and the like.
  • DB database
  • the maintenance of medical equipment delivered to medical institutions is carried out by the medical equipment manufacturer that delivered it, and installation equipment is required for each facility or patient, so a service business for equipment management service provision type is formed. It's getting better.
  • the database used in this embodiment may be the one in this service.
  • This service is configured to centrally manage including customer information and provide configuration management and change management functions for installed equipment based on customer data such as clinical departments, various patient groups, doctors, and nurses. By registering items (configuration items) and sequentially updating the configuration information of installed equipment, it is easier to manage the range of influence when upgrading or replacing parts. Therefore, it is possible to predict the replacement time and maintain the quality of the installed equipment.
  • medical institutions have been described here, as an exception, information on equipment and / or equipment and / or environment at the time of a health response event, including medical treatment at medical institutions, is also described above. It is assumed that it will be centrally managed within the service of.
  • a health response event is an event that has an impact on health. For example, a user goes to a medical facility to see a doctor, or purchases and takes medicine at a pharmacy, which is related to health. It is an act to do. In addition, users went to the gym for training, exercised tennis, etc., and also had binge eating at restaurants, lacked sleep, and worked or played in the cold. Any act that directly or indirectly affects health, such as.
  • an inference engine may be used to determine the information to be transmitted to the target person by inference.
  • the inspection data of the target person is input to the inference engine, and the information transmitted to the target person is acquired.
  • the reliability of the inference result is determined, and if the reliability is low, teacher data is collected and learning for generating an inference model is requested (see, for example, S8, S10, etc. in FIG. 7).
  • Teacher data is created by obtaining data similar to the data input to the inference engine. When a new inference model is generated, this inference model is used to determine the information to be transmitted to the target person.
  • the time-series test data changes (see, for example, FIG. 8 (a)). Impact events including such environmental changes can be inferred from the user's inspection data by generating an inference model using the time-series inspection data of other users and the like. .. That is, if test data for mainly inspecting biological information is acquired in a time series for a specific period using a test device having a specific specification and the changes are arranged in a time series, the test data changes every moment. A change pattern is obtained.
  • the tendency of the change of the health condition or the physical condition of the person can be obtained by observing the change of the numerical value. If this value is constant over time, it is considered that the health condition is kept constant from the value. In addition, if there is a gradual change due to the rhythm of life, the influence of the season or climate, growth, aging, etc., it may be healthy, so it does not matter. That is, the time-dependent change pattern of the inspection data is gradual or regular, but is substantially constant.
  • the measured value will change in some way. If what is written as a pattern here is a continuous minute change in data, the expression that the pattern changes means that the measured value has a tendency (trend) even though it changes slightly, that the trend is changing, and that the trend is changing. When the above-mentioned pattern of meaning is changing, it is confusing if the two expressions are not distinguished. Therefore, if there is a cycle in which the fluctuation is small, but it rises or falls, and there is a cycle, the term that expresses the situation in which it changes from normal may be rephrased as a trend change.
  • the pattern small fluctuations centered on a specific value
  • the trend is also constant.
  • the numerical value may be different from the previous tendency, or it may fluctuate depending on the change slope or fluctuation range, and the tendency (trend) up to that point may be changed. If it does not return (if it changes to a different trend due to a peculiar change), it is probable that the person had some change in health or physical condition. At this time, it is assumed that the trend of the pattern has changed.
  • the tendency of such a change pattern (this is a steady change) is monitored, and the tendency (maximum value, minimum value, average value, fluctuation cycle, etc.) in a specific period such as one day, one week, or one month unit. Record and understand the tendency of fluctuation range). Then, the newly obtained test results may be collected into a pattern, and the individual test values of this pattern may be compared to determine whether the tendency of this pattern has changed the tendency (trend) so far. Of course, if a numerical value with a significantly different trend appears, it is clearly considered to be different from the trend. Further, it is not necessary to make a single comparison, and it may be determined by the tendency of the pattern of a cohesive data group, such as when the tendency change continues.
  • the event when there is a change in the tendency of the change pattern affects the physical condition and health, so it is tentatively called an influence event.
  • the trend of the change pattern changes within a specific period (in a broad sense, it may be expressed that the pattern has changed, so in order to avoid complication, the trend of the pattern has changed during the following examples.
  • Detecting the impact event information at the time corresponding to (sometimes expressed as a pattern change) (you may search for the recorded one or search for external information at that time) is the person. It can be important information for health management not only for other people but also for other people.
  • the result of annotating the detected influence event information can be obtained as teacher data used for inference model learning for event inference for health management.
  • An inference model trained from such teacher data is acquired, and the inspection data obtained in time series in a period similar to the width of a specific period using an inspection device of a specific specification of a specific person is used as an inference model.
  • the reason why I wrote the width of a specific period here is that the change in data obtained over many years and the change in data obtained in one week are not correct comparisons when comparing test data between the same people. So, in the case of numerical values that change in the morning, day and night, when waking up, when exercising, etc., even the characteristics of the change pattern (small fluctuation centered on a specific value) that considers those effects are compared as a pattern that correctly captures To do.
  • the above period may be selected based on the following concept.
  • the above-mentioned similar period at the time of information acquisition follows the recommended specifications of the database and inference model. It is a period corresponding to a specific period to be decided, and waits until future test results are collected.
  • the above-mentioned similar period at the time of information acquisition has already been acquired by the person who has already received the information, or a caregiver or a person concerned.
  • Time-series data of a period similar to the period corresponding to the period (and may be selected with a similar tendency) is selected from the database and used. Also, select an inference model that uses the data of such a period, or create a new inference model by converting the data of a similar period in the database into teacher data and use it.
  • the timing that I want to know now for example, when I wake up or before going out and I have to decide what to do in the future
  • the complementary information as described above may be used.
  • the period is set to be approximately the same.
  • the database assumed here is assumed to include event information that the trend of the change pattern of the test data changes depending on the event, and even if the database is divided for each event that affects health. Good.
  • event information that the trend of the change pattern of the test data changes depending on the event, and even if the database is divided for each event that affects health.
  • the trend change of the test result and its cause can be estimated, and the person who receives the information can pay attention to the maintenance of health and the prevention of improvement and deterioration.
  • This information transmission system includes a control unit 1, an information determination device 2, a terminal 4, a learning unit 5, an inference engine 7, a database (DB) unit 8, and related inspection institutions (including medical institutions) 9. It should be noted that this database may be mediated by a plurality of linked databases and computers.
  • the control unit 1 is arranged in the server, and the information determination device 2, the terminal 4, the learning unit 5, the inference engine 7, the DB unit 8, and the related inspection organization 9 are connected to the server via a network such as the Internet. It is possible to connect.
  • control unit 1, the information determination device 2, the learning unit 5, the inference engine 7, and the DB unit 8 may be a server. It may be arranged inside, and the others may be arranged in another electronic device such as a server or a personal computer. Further, the related inspection agency 9 may have a server function.
  • the control unit 1 is a controller that controls an information transmission system according to the present embodiment, and is a CPU (Central Processor Unit), a memory, and an HDD (Hard Disc) that provide files, data, and the like to a server and the like and other terminals via a network. It is assumed that the IT equipment is composed of Drive) and the like. However, the control unit 1 is not limited to this configuration, and when it is constructed as a small-scale system, it can be configured with something like a personal computer.
  • the control unit 1 has various interface circuits, can cooperate with other devices, and can perform various arithmetic controls by a program.
  • the control unit 1 receives information from each linked device, organizes the information, generates necessary information, and provides this information to the user.
  • the control unit 1 also has a function of outputting a request to each of the linked devices and operating each device.
  • the information determination device 2 the terminal 4 and the like owned by the target person, and the control unit 1 by wireless communication or wired communication. ..
  • a wireless LAN or a mobile phone communication network is assumed, and short-range wireless communication such as Bluetooth (registered trademark) or infrared communication may be used in combination depending on the situation.
  • the description of the communication unit including the communication circuit, the antenna, the connection terminal, etc. is complicated, so it is omitted in FIG. 1, but the communication unit having the communication circuit, etc. is provided in the part of the arrow indicating the communication in the figure. Has been done.
  • the control unit 1 includes a communication control unit 1a, an ID determination unit 1b, an information provision unit 1c, an inference model specification determination unit 1d, an inference request unit 1e, and a search unit 1f.
  • Each of these parts may be realized by software by a CPU, a program, or the like in the control unit 1, may be realized by a hardware circuit, or may be realized by coordinating software and a hardware circuit. You may. Further, in FIG. 1, since each unit in the control unit 1 cooperates with each other to perform each function, the direction of the signal is omitted, but this will be described separately with a flowchart. For example, in a step like S1 in FIG. 4, the ID determination unit 1b collects information from the information determination device 2 and the like for each of the same target persons.
  • the communication control unit 1a has a communication circuit and the like, and includes an information determination device 2, a terminal 4, a learning unit 5, an inference engine 7, a database (DB) unit 8, and a communication unit provided in the related inspection organization 9. Send and receive data, etc.
  • Each device / part such as the information determination device 2 and the terminal 4 also has a communication unit, but the illustration is omitted in FIG. 1 because it is complicated.
  • the ID determination unit 1b collects information from the information determination device 2 and the like for each of the same target persons. An ID is assigned to each individual in order to identify the individual whose information has been acquired by the information determination device 2. In the present embodiment, since the data of each user is handled, the ID determination unit 1b manages which user's information is received and which user is given the guide. The determination of the specific user is that the information determination device 2 has a biometric authentication function, the user communicates with the terminal 4 through the communication unit in the information determination device 2, and the terminal 4 reads a unique code. Do by. In addition, in order to protect personal information, necessary parts are encrypted and management is strict, but since these are general-purpose technologies, detailed description will be omitted.
  • the information providing unit 1c has a function of acquiring user information (may refer to the result acquired by another device) in order to provide correct information to the user.
  • the information providing unit 1c includes inspection data of the user (specified by the ID) acquired from the information determination device 2 and the like, various information acquired from the related inspection organization 9, and a possessed device stored in the DB unit 8.
  • this information is provided to the user.
  • the information providing unit 1c provides the user with information for recommending a facility suitable for receiving an examination or treatment.
  • the information providing unit 1c inputs the inspection data transmitted from the information determination device 2.
  • this data is inspection data (time series information) to which time information is attached, and is stored in the DB unit 8 or the like with a data structure that can be graphed as shown in FIG.
  • the control unit 1 provides information to the user using the information from the information determination device 2, but the server having the related inspection organization 9 collects the information in the same manner. It may be a modified example such that
  • the information providing unit 1c may acquire lifestyle habits such as the user's address, behavioral style at the place of work, eating habits, bedtime, and meal timing on the Internet, and the acquired information is also taken into consideration. Therefore, information such as facilities to be provided to the user may be generated. The acquisition of this information can be complemented by general-purpose or well-known technology.
  • the information providing unit 1c may also customize the information of the facility or the like generated by acquiring the information. Specifically, it is conceivable to provide information on a clinic near the user's residence (recorded as the user's profile information). However, if the clinic does not have the essential inspection equipment, it is not possible to estimate the cause of the disease, take countermeasures, or treat it. Therefore, we will provide information by adding profile information related to the facilities of the clinic, the specialty of the doctors who work, and other facilities. Profile information about this facility is obtained as medical institution information from the related inspection institution 9.
  • the information providing unit 1c uses the information on the possessed equipment, etc. stored in the DB unit 8 in addition to the information collected from the information judgment device 2 and the related inspection organization 9.
  • the information recorded in the DB 8 may be recorded in a different recording unit other than the DB unit 8. In this case, there are a plurality of DB parts in FIG. 1, but they are omitted because they are complicated.
  • the information providing unit 1c collects various kinds of information in providing the information. That is, the information providing unit 1c functions as an acquisition unit for acquiring the test data of the target person, the profile information of the target person, and the possessed device information for each test / medical institution.
  • the information providing unit 1c determines whether or not there is specific information in the user's inspection data transmitted from the information determination device 2 and the related inspection organization 9, and if the specific information is detected, further inspection is performed. (See S1, S3, S13 in FIG. 4, S1, S3, S13a in FIG. 7).
  • the specific information is information related to a disease, for example, a numerical value having a difference from a healthy state or a trend (change) of a change pattern (a minute change centered on a specific value). If the value of the specific information itself deviates significantly from the standard value, or if the trend change of the change pattern is remarkable, there is a high possibility that there is some kind of disease.
  • the specific information may be information that can be suspected of a specific disease.
  • the facility recorded in the DB section 8 may be searched when presenting the recommended facility (see FIG. 4).
  • the DB unit 8 is constructed with a database of necessary tests for each specific information or disease, and a database of medical facilities, testing institutions, etc. having testing equipment / equipment for performing these tests. Just leave it.
  • an inference engine may be used to infer advice to the user (see FIG. 7). In this case, when the user's history data or the like is input, an inference engine in which an inference model for inferring the user's disease, necessary tests, recommended facilities, etc. may be used may be used.
  • the information providing unit 1c functions as a transmission information determination unit that determines the transmission information to be transmitted to the target person according to the test data of the target person, the profile information, and the possessed device information for each test / medical institution (S7 in FIG. 4). , S9, S11, S13, etc.).
  • the transmission information determination unit inputs the change pattern (trend in which minute fluctuations centered on a specific value change while shifting the value) information of the target person extracted in a specific time width of the target person into the inference model. As a result, an inference result is obtained, and the transmission information is determined based on the obtained inference result (see, for example, S9 in FIG. 4).
  • the transmission information determination unit determines information on the recommended medical institution for undergoing the necessary examination, or information on the timing when the subject visits the examination / medical institution as transmission information (Fig.). 4 S7, S9, S11, S13, timing Tc and the like in FIG. 3A).
  • the transmission information determination unit described above inputs the change pattern of the inspection data acquired by the inspection data acquisition unit (for example, the information determination device 2) into the inference unit (for example, the inference engine 7), and uses the inference result of the inference unit as the inference result.
  • the transmitted information is determined based on this (see S35 in FIG. 5).
  • the transmission information determination unit is the transmission information that determines the transmission information at a time point after the extraction of the time-series pattern to be transmitted to the target person.
  • the transmission information determination unit extracts the change pattern of the inspection data in a specific time width, inputs the extracted change pattern to the inference unit, and acquires the inference result from the inference unit (for example, FIGS. 3A and 3B). ), See S35 in FIG. 5).
  • the transmission information determination unit described above makes an inference by the inference unit when the change pattern of the inspection data acquired by the inspection data acquisition unit is within a specific range (see, for example, S27 and S35 in FIG. 5). ..
  • the transmission information determination unit does not perform inference by the inference unit when the change pattern of the inspection data acquired by the inspection data acquisition unit is outside the specific range (see, for example, S27 and S29 in FIG. 5).
  • the transmission information determination unit makes an inference by inputting a change pattern of the inspection data of the target person into the inference unit, and determines the transmission information to be transmitted to the target person based on the inference result (for example, S35 in FIG. 5). reference).
  • the information transmission determination unit inputs the time-series change pattern of the inspection data of the subject into the inference unit, the inference unit makes an inference, and based on this inference result, the time-series pattern to be transmitted to the subject is later than the time point. Determine the information to be transmitted in.
  • the inspection data acquisition unit acquires inspection data that is a time-series pattern of the target person for a specific period.
  • This acquired time-series pattern is composed of individual inspection data acquired by measurement at a plurality of different timings, not simply data obtained by one-time measurement, and even changes in the inspection data pattern are used as information. ..
  • a time-series pattern consisting of multiple inspection data it is less susceptible to errors caused by changes in the measurement environment and conditions.
  • it infers the health condition from the end of the specific period to the future period (when the specific period is extended), and makes it possible to predict the future.
  • teacher data can be created by adding the timing information of the subject's examination / visit to the medical institution as annotation information to the acquired time-series pattern. If there is an inference part that has an inference model generated by learning using this teacher data, what is the timing (when the specific period is extended) beyond the specific period (period for acquiring the time series change pattern)? Can be inferred if When generating the inference model used here, the specifications of specific input / output information are specified and learning is performed.
  • the time-series change pattern of the test data of the subject is input to the inference unit, the inference unit makes an inference, and based on this inference result, the transmitted information at the timing from the specific period to the future is transmitted.
  • a transmission information determination unit for determining is provided. Therefore, it is possible to provide a system, an apparatus, a method, a program, or the like capable of transmitting the prediction information at the timing after the inspection acquisition of the time series pattern.
  • the reliability of the inspection data of the subject will be reduced if there is a difference in mechanical performance for each inspection device. Therefore, a large amount of change pattern information of inspection data may be acquired by using the same type of inspection equipment (inspection equipment having specific specifications) and treated as big data.
  • the specific period does not have to be a fixed period, and may be a different time width (specific period) depending on the situation.
  • the "time width” here is not the time between measurement timings (inspection interval / measurement interval), but the time interval from the first measurement to the last measurement when acquiring a series of inspection data. Say.
  • the "time width” may be rewritten as a specific time width including a lot of time series data in this time width and change pattern information of the inspection data.
  • the information providing unit 1c which functions as a transmission information determining unit in the present embodiment, inputs a change putter of inspection data into the inference engine 7 in which the inference model generated by the learning unit 5 is set, and infers the result regarding advice. Is obtained and provided to the user corresponding to the input inspection data.
  • This service may use personal information, and may require a contract for personal information in order to receive advice. In that sense, the user's profile information may be important.
  • advice may be delivered to a person who takes care of the user, a caregiver, or the like. In this case as well, valid information such as advice arrives according to the information managed by the user's profile information.
  • the information providing unit 1c functions as a data acquisition unit that acquires inspection data in a time series of a specific period using an inspection device having a specific specification and acquires change pattern information of the inspection data (for example, FIG. 8 (b), see S61 and S63).
  • the information providing unit 1c responds to the input of a specific health response event, and functions as a data acquisition unit that acquires and records inspection data in chronological order using an inspection device having specific specifications prior to the health response event ( For example, see S61 and S63 in FIGS. 8 (a) and 8 (b)).
  • the information providing unit 1c functions as an information transmitting unit that transmits the result of inference.
  • the information providing unit 1c functions as an information transmission unit that transmits a health response event customized for a specific person from the inference result by the inference unit to the specific person.
  • the information providing unit 1c functions as a data acquisition unit that acquires a plurality of inspection data in a time series over a specific period using an inspection device having a specific specification.
  • the information providing unit 1c uses a database to extract events that affect the trend change from the event information based on the relationship between the trend change of the change pattern and the timing of the event, and makes it possible to provide the information. Functions as.
  • the inference model specification determination unit 1d determines the specifications of the inference model to be generated when the inference request unit 1e requests the learning unit 5 to generate the inference model.
  • the control unit 1 acquires the biometric information of the user from the information determination device 2 and the like, and accumulates the biometric information.
  • the control unit 1 requests the learning unit 5 to generate various inference models using the accumulated biological information as teacher data.
  • the inference model specification determination unit 1d determines what kind of specification the inference model is requested in generating the inference model. For example, as shown in FIG. 3A, which will be described later, when time-series biometric information is accumulated, what kind of inspection data (value) does the inference model specification determination unit 1d have, and what is the user?
  • the inference model specification determination unit 1d determines specifications for generating an inference model that infers a facility recommended for receiving further necessary tests and treatments based on time-series biological information.
  • the inference request unit 1e requests the learning unit 5 to generate an inference model of the specifications determined by the inference model specification determination unit 1d. That is, the inference requesting unit 1e requests the learning unit 5 to generate an inference model when a predetermined number of biological information acquired by the information determination device 2 is accumulated, and receives the generated inference model. This received inference model is transmitted to the inference engine 7.
  • the control unit 1 may prepare a plurality of inference models and appropriately select the inference model according to the information to be provided to the user.
  • the search unit 1f is an inspection institution or medical institution having equipment necessary for the inspection or treatment when it is found that further examination or treatment is necessary based on the biometric information of the user acquired by the information determination device 2. Is searched in the database stored in the DB unit 8.
  • the search unit 1f functions as a search unit for searching for an event that affects the trend change in the event information based on the relationship between the trend change of the change pattern and the event timing using the database.
  • the information determination device 2 is a device for acquiring test data such as health-related information of the target person, for example, vital information and sample information.
  • health-related information there are various kinds of information, for example, vital information such as body temperature, blood pressure, and heartbeat of the subject.
  • sample information such as excrement such as urine and stool of the subject, sputum and blood.
  • the information determination device 2 acquires the color, shape, amount, and date / time information.
  • the information determination device 2 may acquire information according to an instruction from the control unit 1, may acquire information according to a user's operation, or may automatically acquire information.
  • the information determination device 2 is used for daily life such as daily life, work / school activities, meals, sports activities, etc. in the information "Personal Health Records (PHR)" which is medical / health information. Personal Life Records (PLR), which includes various activity data of the above, may be collected and utilized.
  • PHR Personal Health Records
  • the acquired information is controlled through a communication unit (not shown) in the information judgment device 2. It is transmitted to part 1.
  • the information determination device 2 functions as an inspection data acquisition unit that acquires inspection data of the target person.
  • the information determination device 2 functions as an inspection data acquisition unit that acquires inspection data that is a time-series pattern of the target person for a specific period.
  • the inspection data of the target person acquired by the inspection data acquisition unit is obtained by acquiring inspection data in chronological order using an inspection device having a specific specification and extracting change pattern information of the inspection data in a specific time width. is there. That is, as the information judgment device 2, an inspection device of a specific specification (inspection device of the same type) is used, and the information judgment device 2 measures the inspection items of the same subject at different timings in chronological order. Get the data to.
  • a change pattern can be obtained by drawing the measured values on a graph according to the inspection timing.
  • Inspection data can be obtained by extracting this change pattern in a specific time width.
  • the inspection data is based on a color sensor for defecation, a shape sensor, a hardness sensor, an olfactory sensor (including reaction judgment of nematodes and animals), a gas component sensor, a color change detection sensor when a specific reagent is added, and a magnified observation image. It is the data obtained according to one of the output results of any of the shape determinations.
  • a wearable terminal When a wearable terminal is used as the information determination device 2, vital information such as body temperature, heartbeat, blood pressure, brain wave, line of sight, respiration, and exhalation can be obtained by closely contacting the skin or the vicinity of the body depending on the wearing part of the wearable terminal. Is possible.
  • a sphygmomanometer As a scale, a sphygmomanometer, and a measuring instrument for measuring arterial stiffness, which means the hardness of the arterial wall, dedicated precision equipment is installed in health facilities, public baths, pharmacies, shopping malls, etc. The measurer may also be assigned. In such facilities, users often use the measuring device comfortably in their spare time and manage their physical condition based on the measurement results at that time. These measuring devices may be used as the information determination device 2.
  • the information judgment device 2 may request the user to fill out a questionnaire before and after using a dedicated terminal or the like. In such cases, the user's profile information and other information can be identified based on the description in this questionnaire. Such information collection is not limited to the information determination device 2, and may be performed by the control unit 1. This information can be used when determining whether or not the specific information has been acquired in step S3 of FIG. 4, which will be described later. If information such as when the doctor went to the doctor can be heard, it can be used as the time Tc information in FIGS. 3 (a) and 3 (b) described later.
  • the information determination device 2 obtains information on a specific user
  • the information on the facility recommended by the information providing unit 1c of the control unit 1 is presented to the information terminal 4 of the specific user.
  • the explanation is given on the assumption that this presentation assists the user's behavior, but various variations can be considered.
  • the information determination device 2 may be a thermometer or a sphygmomanometer that is already suffering from a specific disease and is used under the guidance of a doctor.
  • the mobile terminal (smartphone) provides the information as it is. It can be the judgment device 2.
  • the extent to which the information determination performed by the information determination device 2 or the like is determined may be changed in relation to the control unit 1.
  • the information determination device 2 or the like may transmit the sensing result to the control unit 1 without determining only the result.
  • the attached information is associated with which person and which sensing result, but it may be associated with the information of another terminal by adding it.
  • the related inspection institution 9 is a facility where the user is inspected, and there are, for example, an inspection facility and a medical facility.
  • the related inspection organization 9 may be a mobile type, for example, a type in which general medical equipment or inspection equipment is mounted on an automobile, train, ship, helicopter, drone, or the like and the patient goes to the patient.
  • the control unit 1 can acquire which medical institution the patient went to, what kind of test result was obtained, and the like from the server that operates the system of the related test institution 9.
  • the server of the related inspection organization 9 may be the same as that of the control unit 1, or some functions may be shared.
  • the related testing institution 9 in the above description, a medical-related test using specific biological biological information or a sample sample from a living body is described.
  • questions and answers such as interviews and subjective symptoms are also tests
  • one of the inspection institutions is one that produces some results by inputting in natural language based on the questions and answers. Be done.
  • the Internet service that provides specific health advice by questionnaire description or item selection is also included in the broad interpretation of the related inspection organization 9.
  • the related inspection organization 9 inputs information corresponding to the category organized and linked for each specific area of the displayed screen, database search and logic-based branching are performed for each item content.
  • inference processing it is possible to provide a service using a technique of displaying health-related advice in characters or by voicing it to the user by voice.
  • the terminal 4 is a mobile information terminal, and is a device for receiving information that can be confirmed by the target person and related persons.
  • the terminal 4 may be, for example, a smartphone or a tablet PC.
  • the built-in camera or microphone can be used as an information acquisition unit.
  • a wearable terminal or other home appliances that can be linked may be used as the terminal 4, and information may be acquired by the wearable terminal or the like. Therefore, the information determination device 2 and the terminal 4 may be the same device, or may be dedicated devices, respectively. Further, depending on the situation, the information determination device 2 or the terminal 4 may have the function of the control unit 1, and may be configured to share detection, control, and information provision.
  • the database (DB) section 8 is a database of information on devices owned by medical facilities and the like. Owned equipment includes measuring equipment used for diagnosis, equipment / equipment for performing treatment, and the like.
  • the control unit 1 refers to the information recorded in the database unit 8 to provide the information regarding the recommended facility.
  • the DB unit 8 has an electrically rewritable non-volatile memory.
  • the DB unit 8 has a recording unit 8a that records a list of owned devices for each facility, and a recording unit 8b that records an ID and a user's visit for each facility.
  • the recording unit 8a records a list of devices owned by facilities such as hospitals, clinics, and inspection institutions.
  • the information providing unit 1c can present the information of the facility having the most suitable equipment for the inspection to the user.
  • the information may be linked with the information of the related inspection institution 9.
  • the recording unit 8b records visit information indicating which person (identified by ID) came at what time for each facility.
  • the recording unit 8b may be omitted.
  • the DB unit 8 is constructed as a part of an information transmission system in which medical facilities cooperate, and the DB unit 8 may also be able to access the related inspection institution 9 through the control unit 1. In this case, when the DB unit 8 receives a search command from the control unit 1, the DB unit 8 searches the data in the related inspection organization 9 in addition to the data recorded in the DB unit 8 to search. Output the result.
  • the DB unit 8 functions as a storage unit that stores the profile information of the target person and the possessed device information for each examination / medical institution.
  • the storage unit is not limited to the DB unit 8, and all or part of its functions may be arranged in the control unit 1 or the like. Details of the data recorded in the DB unit 8 will be described later with reference to FIG.
  • the DB unit 8 includes the change pattern information with specific improvement timing information such as taking medication and the start of life improvement as time information similar to the time information of the change pattern information, so that the part of the change pattern before the improvement information is included. It functions as a data creation unit that creates a database with comparable information.
  • the DB unit 8 creates a database that can record the change pattern information of the inspection data created by acquiring the inspection data and the timing of each of the plurality of event information regarding the acquisition source of the inspection data within a specific period in association with each other. It functions as a data creation unit to be created.
  • the learning unit 5 has an input / output modeling unit 5a and generates an inference model by machine learning or the like.
  • This inference model is generated by learning the relationship between acquired information such as acquired biometric information and biopsy information and the disease, and specifically by learning the relationship between the acquired information and the clinical department / department.
  • the input / output modeling unit 5a Similar to the inference engine 7, the input / output modeling unit 5a has an input layer, a plurality of intermediate layers, and an output layer, obtains the strength of the connection of neurons in the intermediate layer by learning, and generates an inference model.
  • the learning unit 5 detects the influence event information at the time corresponding to the time when the change pattern changes within a specific period with respect to the change pattern information, and learns the result of annotating the detected influence event information as teacher data. It functions as a learning unit for acquiring the inference model (see, for example, S65 to S71 in FIG. 8). In addition, the learning unit 5 annotates the change pattern information of the recorded test data with information on the equipment, / or equipment, and / or environment at the time of the health response event as teacher data. It functions as a learning unit that acquires an inference model trained as (see, for example, S65 to S71 in FIG. 8).
  • the change pattern of the test data acquired from the subject using the same type of test device is extracted in a specific time width, and the extracted change is extracted.
  • the pattern is input to the inference engine 7, and the teacher data is generated using the health advice to be output at a later timing as the annotation information from the timing of the examination by the subject.
  • an inference model is generated by performing learning using this teacher data.
  • the time width that goes back from the time when the test result is obtained has been described, but if the data improves after the test result due to treatment or the like, the treatment may not be successful. You may learn and output prognostic advice.
  • the learning unit 5 can generate an inference model capable of providing lifestyle-related improvement and future prediction advice on the effects of treatment and medication by learning using the examination data sequence after examination, hospital visit, and medication. You can also do it.
  • the time-series data is used starting from the time of examination, outpatient visit, and medication. Use the previous time-series data when giving advice on tests, hospital visits, medications, etc.
  • Deep learning is a multi-layered structure of the process of "machine learning” using a neural network.
  • a typical example is a "forward propagation neural network” that sends information from front to back to make a judgment.
  • the simplest forward-propagating neural network is an input layer consisting of N1 neurons, an intermediate layer consisting of N2 neurons given by parameters, and N3 corresponding to the number of classes to be discriminated. It suffices if there are three layers of output layers composed of the above neurons.
  • Each neuron in the input layer and the intermediate layer, and each neuron in the intermediate layer and the output layer is connected by a connection weight, and a logic gate can be easily formed in the intermediate layer and the output layer by applying a bias value.
  • the neural network may have three layers as long as it makes a simple discrimination, but by increasing the number of intermediate layers, it is possible to learn how to combine a plurality of features in the process of machine learning. In recent years, those having 9 to 152 layers have become practical from the viewpoints of learning time, determination accuracy, and energy consumption.
  • a "convolutional neural network” that compresses the feature amount of the image, performs a process called “convolution”, operates with the minimum processing, and is strong in pattern recognition may be used.
  • a “recurrent neural network” (fully connected recurrent neural network) that can handle more complicated information and whose meaning changes depending on the order or order may be used.
  • a conventional general-purpose arithmetic processing circuit such as a CPU or FPGA (Field Programmable Gate Array) may be used.
  • a processor called GPU (Graphic Processing Unit) or Tensor Processing Unit (TPU) specialized in matrix calculation is used.
  • GPU Graphic Processing Unit
  • TPU Tensor Processing Unit
  • NPU neural network processing unit
  • AI artificial intelligence
  • machine learning examples include, for example, support vector machines and support vector regression.
  • the learning here is to calculate the weight of the discriminator, the filter coefficient, and the offset, and there is also a method using logistic regression processing.
  • humans need to teach the machine how to make a judgment.
  • a method of deriving the judgment of the image by machine learning is adopted, but in addition, a rule-based method of applying the rules acquired by humans by empirical rules / heuristics may be used.
  • the inference engine 7 has an input / output layer and a neural network similar to the input / output modeling unit 5a of the learning unit 5.
  • the inference engine 7 makes inferences using the inference model generated by the learning unit 5.
  • the inference engine 7 is measured by the information determination device 2, inputs time-series biometric information, and, for example, obtains an appropriate inspection institution / medical institution for inspecting, treating, or the like of the user's health condition by inference. ..
  • time-series biometric information it may be inferred when a medical institution will receive a medical examination.
  • the inference engine 7 functions as an inference unit having an inference model learned according to the timing information of the subject's examination / visit to the medical institution.
  • the inference engine 7 functions as an inference unit having an inference model learned according to the change pattern of the person who provided the examination data similar to the examination data of the subject and the examination / visit to a medical institution / examination / medication information.
  • the inference unit acquires inspection data of a person other than the target person using an inspection device of a specific specification, extracts the change pattern of the inspection data in a specific time width, and uses it as an input of the inference unit for learning. Inference using the inference model generated by the learning inference unit using the health advice for the subject to be output from the end of the time width measured by the examiner to a later time as annotation information. I do.
  • the inference engine 7 inputs the inspection data obtained in time series in a period similar to the width of the specific period using the inspection device of the specific specification of the specific person into the inference model to infer the influence event. It functions as an inference unit (for example, inferring when transmitting the inspection results of FIGS. 4, 7, etc. using the inference model generated in FIG. 8 (b)).
  • the inference engine 7 inputs inspection data obtained in time series using an inspection device of a specific specification of a specific person into an inference model to infer information on equipment and / or equipment and / or environment. It functions as an inference unit (for example, using the inference model generated in FIG. 8 (b), inference is made when transmitting the inspection results of FIGS. 4 and 7).
  • the inference model learned according to the subject's visit / examination / medication information is an inference model in which the input / output relationship is set by the teacher data annotated with the input as time-series examination data and the output as related disease information.
  • the related disease information is the date and time of the visit, the diagnosis result, the prescription drug information, and the like.
  • control unit 1 may provide information on the recommended facility by using the inference engine 7 in addition to the search unit 1f searching the DB unit 8.
  • the inference engine 7 infers information about the recommended facility using the inference model generated by the learning unit 5.
  • This inference model is generated by learning the relationship between acquired information such as acquired biometric information and biopsy information and the disease, and specifically by learning the relationship between the acquired information and the clinical department / department. In this way, the control unit 1 may output the guide information to be presented by the inference by the inference engine 7.
  • control unit 1 guides medical facilities, etc. with a single judgment based on the acquired information obtained at one time by search or inference, it unnecessarily brings medical information into life and lives soundly and with peace of mind. May interfere with. Therefore, the accuracy may be improved by using the history (time-series information) of the acquired information a plurality of times.
  • FIG. 3A shows a graph using personal health-related historical data (time-series data) recorded in the recording unit 8.
  • the recording unit 8 records, for example, the specific data A among the data acquired by the specific device A or the data of the device having various inspection functions, and when and which facility the individual visited. doing.
  • the control unit 1 manages the recording of the recording unit 8.
  • the horizontal axis of the graph shown in FIG. 3 (a) is time, and the vertical axis is health-related data.
  • the data posted on this graph can be treated as if the information were arranged two-dimensionally. Therefore, inference can be made by a method similar to image search, in which a specific object is found from this image. That is, the graph may be the input and the output may be the medical institution information. Details of the graph showing the historical data shown in FIG. 3 will be described later.
  • the recording unit 8 records what kind of facility the person whose health-related information has changed is going to or is going to, and when the control unit 1 centrally manages this record, these data are collected. It is possible to have the learning unit 5 create an inference model by using it as teacher data.
  • the inference engine 7 has an inference model obtained by inference model specification determination unit 1d of the control unit 1 designating an inference model specification and learning according to this specification.
  • the inference engine 7 may have a plurality of inference models because the device data may be different when a new device appears.
  • new learning is required each time a new device appears, it is assumed that the inference model is often improved or newly created through the learning unit 5 by the designation of the control unit 1.
  • the information determination device 2 is dedicated and specialized only in inference of a specific disease, it may be a single dedicated inference model.
  • the inference engine 7 is a circuit block centered on an AI chip such as a CPU, GPU, and DSP, like the input / output unit 5a of the learning unit 5, and also includes a memory and the like to form a neural network.
  • the inference engine 7 and the learning unit 5 are connected to a network or the like operated in cooperation with a hospital or the like, and the control unit 1 may be used in cooperation with these. I'm assuming. In this case, there is a possibility that learning and inference information can be exchanged via the related inspection organization 9.
  • the inference engine 7 When the inference request unit 1e of the control unit 1 determines that the information of a specific user has been sufficiently acquired from the information determination device 2 or the like, the inference engine 7 is requested to make an inference.
  • the inference engine 7 inputs an information group showing the time-series transition of similar data, infers based on this information group, and outputs medical institution information (visit information, examination information, etc.) suitable for a specific user. It is possible to do. If a group of data similar to a person with a chronic medical condition is input to the inference engine 7, it is better to display a guide to a facility where the same treatment can be performed.
  • the inference engine 7 can estimate the medical institution that has the equipment to inspect a specific user.
  • the table of FIG. 2 shows the temporal changes and transitions of the data output by the information determination device 2 (test data, biological data, vital data, sample data, etc. are generally expressed as "time series data"). (Explanation as patient ID) indicates what kind of medical institution, etc. was associated with. From this table, it is possible to know when the patient came to any hospital or clinical department based on the patient's behavior history and the history of the in-hospital system.
  • the hospital data recorded in the DB department 8 includes what kind of clinical departments there are, what kind of illnesses have been treated, and what kind of doctors, nurses and other staff are working. Information such as the timetable and other consultation conditions, including what kind of examination can be taken, how long the waiting time is, and whether it is a reservation system or a referral system, is organized and recorded.
  • the consultation conditions may include information such as what kind of questionnaire must be filled out.
  • the consultation conditions may include not only information related to consultation, hospital visit, and surgery, but also information on related rehabilitation facilities, prognosis visit, treatment information, and the like.
  • recording data in a tabular format is possible not only in hospitals but also in pharmacies and public facilities when undergoing examinations.
  • pharmacies there may be information such as what kind of medicines are prepared and provided.
  • the medical institution information is organized including the above-mentioned information and the characteristics of the attached pharmacy, it will be easy to visit in the neighborhood regardless of the medical institution specializing in the treatment of a specific medical condition. You can substitute it by going to your own medical institution. It is possible to select and visit an appropriate medical institution according to the characteristics of the disease, beyond the characteristics of the clinical departments that each hospital / clinic specializes in or has. Such information needs to be reviewed at an appropriate cycle because there are changes in medical staff who work, retirement, and replacement of equipment.
  • medical institution information is not limited to consultations and outpatient visits, but may include information on non-illness and prognosis, and health-related facilities visited by patients (public institutions / facilities that promote or promote recovery of health, etc.) Or, there may be information on stores that provide food and drink that affects the physical condition by ingestion), and what kind of equipment is available is also displayed in a list.
  • Large hospitals have many facilities, and small clinics have limited facilities. The number of cases related to the disease tends to be higher in large hospitals.
  • a specialized clinic specializing in a specific disease can supplement the environment surrounding a small clinic, and can have the same function as a large hospital by cooperating with the clinic in a smart city or the like. Therefore, here, it is possible to present it to the patient as effective information.
  • Such feature information is also organized in chronological order, and in the table shown in FIG. 2, the data output by the information determination device 2 (test data, biological data, vital data, sample data, etc. are generally expressed as "time series data"). You may treat it by incorporating it into the information in the same way as what kind of temporal change or transition the data has changed.
  • It is a data group that summarizes what kind of biological data changes and what kind of lifestyle (dietary tendency, etc.) patients (explained as patient ID) are related to what kind of medical institution.
  • patient ID a data group that summarizes what kind of biological data changes and what kind of lifestyle (dietary tendency, etc.) patients (explained as patient ID) are related to what kind of medical institution.
  • patient ID a data group that summarizes what kind of biological data changes and what kind of lifestyle (dietary tendency, etc.) patients (explained as patient ID) are related to what kind of medical institution.
  • patient ID a data group that summarizes what kind of biological data changes and what kind of lifestyle (dietary tendency, etc.
  • This information transmission device, etc. does not provide typographical advice such as going to a dermatologist or a digestive system clinic, but has similar characteristics within the range that the person can easily access.
  • the characteristic information of medical institutions, health-related facilities, and stores in the vicinity is updated and renewed at a specific timing, and the situation at the time of the change in biometric information is recorded and can be searched.
  • the items shown as "others" in FIG. 2 may include not only the medical institution alone but also the surrounding situation and their past information. That is, test data is acquired in chronological order using a test device having specific specifications, change pattern information of the test data is acquired, and characteristic information (characteristic information) in a medical institution or the like that has an improvement effect on the change pattern information (The inferred result is transmitted by inputting the inspection data obtained in time series using the inspection equipment of the specific specifications of a specific person to the inference model in which the result of annotating the impact event information) is trained as teacher data. It becomes possible to provide an information transmission method characterized by doing so.
  • inspection data is acquired in chronological order for a specific period using an inspection device having a specific specification.
  • the change pattern information of the inspection data is acquired from the inspection data acquired in time series.
  • the influence event information at the time corresponding to the time when the change pattern changes within a specific period is detected.
  • the impact event information is information on the event that contributed to the improvement when it is recognized that the improvement effect was achieved by the medical institution or the like.
  • an inference model is acquired in which the result of annotating the detected impact event information is trained as teacher data.
  • the inference model When the inference model is generated, the inference model is affected by inputting the inspection data obtained in time series during a period similar to the width of the specific period using the inspection equipment of the specific specification of a specific person. Infer the event. When the inference result is obtained, the inference result is transmitted to the user or the like.
  • the time-series pattern here and the database that records this time-series pattern can be said to have the following characteristics.
  • the database can include an example in which the change pattern information of the inspection data is acquired after the specific improvement timing and the feature that the change pattern is surely improved is also included.
  • the timing of improvement if it is possible to determine the similarity of the change pattern of the patient in the future and it contains some information that the change pattern of the biological information of the patient candidate may be improved, many people will be asked. It can be used and become a convenient database.
  • possession equipment includes simple items such as thermometers, stethoscopes, sphygmomanometers, weight scales, and body fat scales, as well as X-ray examination devices.
  • endoscopic ultrasonography equipment stationary equipment such as CT and MRI, and examination equipment and reagents for various infectious diseases.
  • stationary equipment such as CT and MRI
  • examination equipment and reagents for various infectious diseases.
  • these devices can be identified and managed including them.
  • some endoscopes can and cannot observe special light. As a result, the characteristics, types, limits, etc. of the inspections that the facility can perform are determined.
  • the characteristics of the facility including information such as inventory of owned equipment, supplies, consumables, etc. in the column of owned equipment, you can issue an effective guide based on this information. Since the gender of doctors and examination technicians and the existence of private rooms are also useful information for patients, they may be displayed and guided in a database. Whether the toilets in hospitals and other facilities can obtain vital information, sample information, etc., whether they are properly calibrated and highly reliable toilets, or whether they are facilities that support urinalysis and stool tests, and whether they are vitals. It is preferable to acquire information such as whether or not there is a rental device for monitoring the patient's health condition, which can acquire information, as facility information. Not only patients who come to the clinic, but also users who visit the facility to obtain information on the calibrated toilet at the facility, the error between the toilet at home, and the error between the toilet at home and the stool test can be handled.
  • the following inference model is generated to perform inference.
  • the result of annotating the equipment and / or equipment and / or environment information at the time of the health response event is used as teacher data, and the inference model is acquired by training using this teacher data.
  • the inference model is acquired, as described above, the inspection data obtained in time series using the inspection equipment of the specific specifications of the specific person is input to the inference model, and the equipment and / or equipment and / or By inferring environmental information, the above important factors can be obtained. If you convey a health response event customized for a specific person from this inference result (important factor) to a specific person, you can go to a clinic in the neighborhood without having to go to a distant hospital, or improve your life. It may be done.
  • the system it is possible to make inferences including the elements of health response events.
  • information is not given as to whether or not the person's health condition has really improved, but the actions taken by many people can be inferred.
  • a specific health response event (which may be manual or automatic) may be input only when the health condition is improved. Improvement can be judged by changes in inspection data, etc., or may be judged based on a questionnaire. In other words, the system may be such that the hospital sends a questionnaire mail to the patient after the medical treatment, and the health response event is input only when the answer that the symptom of concern has improved is returned.
  • the event information is subdivided based on the environment and situation of the event that affected the health condition and other component information, it can be provided as customized information replaced by the position of the person who provides the information.
  • test data change pattern and the teacher data of a large number of cases obtained by annotating the influence event may be collected and learned to create an inference model.
  • the input may be the change pattern and the output may be the influence event element.
  • the number of cases Nc1 to Nc6 and the like are recorded in the database of the DB unit 8. If there is a number of cases treated by the clinic for the related disease lp, it can be used as a reference when the user selects a medical facility, and can also be used when displaying a recommended medical facility by inference or the like.
  • information such as prognosis information, hospitalization period information, and adjacent pharmacy information (for example, assortment information may be included) may be recorded in other columns of the database of the DB unit 8. ..
  • the search unit 1f of the control unit 1 has a function of accessing the database as shown in FIG. 2 and extracting necessary information.
  • the patient ID is an identification code given to each patient (user) in the ID determination unit 1b to identify the patient.
  • the time series data is individual data measured by the information determination device 2.
  • the data Dy1 (t11) is the data of the type y1 measured at the date and time t11.
  • the data type y is, for example, vital information such as body temperature and blood pressure, and sample information such as stool.
  • the date and time of visit Th indicates information on the date and time when the patient visited a medical institution.
  • the related disease lp indicates information on the disease name assumed from the time series information of the patient ID.
  • Clinics H1 to H4 indicate medical facilities, and indicate medical facilities where a person specified by a patient ID has visited.
  • the clinical departments Dp1 to Dp3 indicate the names of clinical departments in the medical facility where the person specified by the patient ID visited.
  • the possession equipment Mod is the possession equipment associated with the person identified by the patient ID.
  • the number of cases Nc1 to Nc6 is the number of cases handled in each clinic / clinical department. When displaying the recommended equipment to the user, the number of cases may be taken into consideration, or the recommended equipment may be displayed for reference information when the user selects a clinic or the like.
  • FIG. 2 shows a list in which the recording unit 8a of the list of owned devices by facility and the recording unit 8b of the ID and visit information list by facility are mixed.
  • the recorded contents of the recording unit 8a will be obtained.
  • the recorded contents of the recording unit 8b can be obtained.
  • time-series data is recorded for each patient ID, historical data for each individual patient can be created regardless of the medical facility that received the treatment. This historical data can also be used when making inferences in S9 of FIG. 4 and S6 of FIG. 7, which will be described later.
  • FIG. 3 is a graph created using inspection data.
  • FIG. 2 shows an example of time-series inspection data recorded in the DB unit 8.
  • examination data organized in chronological order is recorded for each patient ID
  • FIG. 3 is a graph showing the examination data shown in FIG.
  • the horizontal axis shows the time T
  • the vertical axis shows the time series data in FIG. 2 plotted.
  • this is test data, biological data, vital data, and sample data, and the test output result of the device that inspects any of these is represented by a numerical value D. For example, it is a value indicating the degree of red color of stool.
  • FIG. 3 it is assumed that the date and time of visit and the like are automatically updated systematically.
  • the example shown in FIG. 3A is a case in which the time-series data changes in the direction of deterioration of health and the user eventually goes to the hospital. Therefore, the pattern of each patient described in FIG. 2 Closest to.
  • the information as explained in FIG. 2 is also used to provide the result of inferring how long and what kind of clinical department the patient will go to. It is possible to do.
  • a table as shown in Fig. 2 is created. It can. However, this does not apply to the table in FIG. 2 because some people have only vital data even if they do not go to the hospital at all.
  • Fig. 3 (a) is a case in which it is presumed that the patient will go to the hospital in the future.
  • the graph shown in FIG. 3A shows changes in time series of examination data (equipment data) of users who are not currently visiting the hospital. From this time-series test data, it is possible to obtain information on whether or not to visit a medical institution when a specific test result (specific information) is obtained. Therefore, based on the time-series test data, it is possible to guide the health information so that the user can grasp his / her health condition by receiving it before it gets worse as he / she goes to the hospital. For example, in FIG.
  • the medical institution in the case of the test data at time T1, it can be inferred that the medical institution is visited at time Tc when time + ⁇ T has elapsed. That is, if the examination data, medical institution information (clinical name, clinical department, date and time information) and the like are accumulated in the DB unit 8, the period until receiving medical treatment at the medical institution can be estimated.
  • Fig. 3 (b) shows a case in which the patient has already visited the hospital and has deteriorated during the hospital visit due to factors other than treatment.
  • the graph shown in FIG. 3B is an example in which a person who goes to the hospital due to illness receives treatment at a clinic when specific information appears at times Tc1 and Tc2.
  • the time-series inspection data as shown in FIG. 3B can be sufficiently utilized to learn such a situation. This example is useful for a guide to the effect that "people with this number usually cannot be treated on their own.” It is effective as information that can prevent further deterioration.
  • Figure 3 (c) shows a case where it is not necessary to go to the clinic.
  • the test data D is lower than the predetermined value (indicated by the broken line in the graph) and there is no need to go to the clinic.
  • the column of the visit date and time is blank.
  • This database may hold the relationship between the type of acquired information (toilet bowl occult blood test information), the clinic, and the owned equipment Mod, and the patient-specific time-series data may be managed in a separate database. Further, by searching a plurality of DBs and organizing the search results, information corresponding to the database recorded in the DB unit 8 may be obtained.
  • FIG. 3 is a graph showing the time-series information for each patient recorded in the DB unit 8, where the horizontal axis is time and the vertical axis is the numerical value of the acquired information. Therefore, the information is two-dimensionally visual. Since it is a two-dimensional diagram, the following two things can be said. First, since it is a diagram, it can be handled in the same way as image judgment, and a general-purpose and easy-to-build AI chip or system such as an inference model for image recognition can be easily diverted, and inference can be easily realized. .. In addition, since the horizontal axis is time, it is possible to effectively use the time-varying information of physical information, and it is possible to easily make a prediction. In addition, since similar data is acquired repeatedly, the number of data is easy to collect, and even if there are some errors in individual data, the tendency is to compare the pattern with other people's data groups to make a judgment. It's easy to do.
  • this time axis it is possible to reflect information on the daily rhythm of daily life events, etc., which will be described later, as well as health and medical aspects that affect or reflect the health of the person. It is good to be able to reflect the event as well.
  • the time when you started exercising is an event on this time axis, and even if you feel better by starting to take some health food, you can check it in relation to this time axis. It can also be associated with changes in data.
  • FIGS. 3 (a), (b), and (c) it is simplified and illustrated in an easy-to-understand manner with an example of whether or not there was a hospital visit. It is possible to statistically judge the pattern change of the inspection data string and the influence of the relationship with these events.
  • inspection data is changed in a positive direction (for example, time-series data group of a corresponding person for a specific period, time-series inspection information, and health-related information is inspected.
  • the data may be added so that it can be identified), and the annotation "good data change" may be performed.
  • an examination data acquisition unit that acquires examination data for each profile
  • a storage unit that stores the medical institution that has been treated for each profile, examination data, profile information, and a medical institution for medical treatment. It has a transmission information determination unit that infers medical institution information and determines the transmission information to be transmitted to the target person by inputting the test data of a specific target person by the reasoning model learned from the information as teacher data. It becomes possible to provide an information transmission device characterized by.
  • the medical institution information used for the teacher data includes features such as the equipment, equipment, and medical records owned by the institution. If a medical institution with a specific facility is inferred, the information may be shared with that specific person, or the medical facility possessing the facility can be used to provide medical care with the device in the vicinity of the person's address. You may search for an institution and use it as information.
  • the time series information can include information on the characteristics of time changes peculiar to the living body, such as fluctuations and frequency. For example, it is easy for the user to consider information such as when sleeping, when waking up, morning and day and night, before meals, after meals, and before and after bathing. In addition, there is a study that it is more relaxed and healthier to have appropriate fluctuations in heart rate and breathing. If necessary, it can be used in various ways, such as extracting only the data at the time when it seems to be the timing as described above, omitting the data in a specific situation, and using only these data.
  • the period during which the history data is acquired corresponds to the specific period, and the inspection data during this specific period is extracted.
  • the extracted time-series inspection data is input to the inference engine 7, the inference engine 7 outputs advice by inference, and the advice is provided to the subject.
  • the specific time width may be suitable for giving some advice at a time corresponding to the future after the end of the time width, and a plurality of information retroactive to the time of this advice is acquired. It should be as wide as possible. Further, the specific time width may be standardized, or may not necessarily be strict, and it may be sufficient if a sufficient amount of data can be obtained.
  • time interval of each data is also important information, it is better that the data is obtained with a regular time width and is not discrete. However, even if the measurement time points of the data are discrete within the time width, it is effective as long as the time width is such that the data can be supplemented by interpolation to obtain meaningful data. It may be determined according to some health or medical information.
  • the information transmission device extracts the change pattern of the inspection data of the subject in a predetermined time width, and determines the transmission information to the subject according to the inference model learned together with the time information. It has a transmission information determination unit.
  • FIG. 4 shows a case where the search in the DB unit 8 shown in FIG. 1 and the inference by the inference engine 7 are individually used. You may want to use one of the functions, or you may use both of them in layers, but here is the simplest example.
  • an image sensor In explaining the flow of FIG. 4, as the information judgment device 2, an image sensor, a magnifying image judgment device such as a microscope, a sensor for detecting special light reflection, an array of crystalline nanowires, and a molecular film are used as the information judgment device 2.
  • the explanation will be made on the assumption that an olfactory sensor, a gas component sensor, etc. that apply changes in electrical characteristics such as the above are arranged and the characteristics of the user's excrement can be confirmed.
  • the recommended facility to be displayed is searched from the database recorded in the DB unit 8 (see S5Yes ⁇ S7).
  • the DB unit 8 is constructed with a database of information on facilities that can perform user examinations and treatments based on specific information.
  • the inference is performed using the user's history data (see S5No ⁇ S9).
  • each ID is determined based on the sensor output result (S1).
  • the control unit 1 may acquire the output of the information determination device 2 through the communication control unit 1a, or the control unit 1 may receive the data transmitted by the information acquisition device 2 in the communication control unit 1a. Further, it is assumed that the data recorded by the information determination device 2 is collected by the control unit 1 through the communication control unit 1a at a specific timing.
  • the inspection result is determined based on the sensor output for each ID attached to the sensor output result.
  • the sensors are a color sensor, a shape sensor, a hardness sensor, an olfactory sensor, a gas component sensor, and a color change detection sensor when a specific reagent is added. Based on the output of the image sensor, the shape is determined by a magnified observation image. May be good.
  • stool with occult blood can be determined by a color sensor.
  • the amount, shape, hardness, etc. of excretion may be determined by an image sensor / color sensor, or a method of measuring the color distribution by performing special dyeing may be used.
  • the composition may be detected in a magnified image of the object, or the result of culturing for a specific time may be determined. For example, when the amount of blood mixed in the stool increases, the red color of red blood cells becomes conspicuous, but if this is quantified, the difference from the healthy case can be seen. These are detected in step S1.
  • step S1 when the control unit 1 determines the sensor output result by a determination using a specific program or the like, it then determines whether or not specific information can be obtained (S3). Further, this specific information may be output by the information determination device 2 after determining that it is "specific information".
  • this specific information may be output by the information determination device 2 after determining that it is "specific information”.
  • it is determined whether or not specific information related to the disease for example, a feature such as a numerical value different from the healthy state is detected.
  • the user's profile, behavior, and lifestyle are determined (S15).
  • the profile, behavior, and lifestyle of the user using the terminal 4 are determined at the timing when the specific information is not acquired.
  • These information of the user may be input by the user from the terminal 4 through the communication control unit 1a by the control unit 1, and the user downloads a profile or the like uploaded on the net such as SNS and accumulates the information. May be good.
  • Such information transfer may be push type, pull type, frequent acquisition, or intermittent acquisition, and the control unit 1 acquires the manual input result of the user to the terminal 4 through the communication control unit 2.
  • it may be stored in the DB unit 8 so that it can be referred to.
  • the communication control unit 1a also intervenes in the communication with the DB unit 8.
  • advice on recommended facilities, etc. is obtained using the biometric information test data for each individual obtained in chronological order. Furthermore, the result of annotating the characteristic information of medical institutions, etc. that had an improvement effect on the change pattern information of the biometric information test data, or the result of annotating the input information of a specific health advice service is learned as teacher data. The case where the inference model is generated and the trigger information at the time of improvement of the inspection data is obtained by using this inference model is described. In addition, by including specific improvement timing information such as taking medication and starting life improvement as similar time information to the change pattern information, improvement of the test data of other people whose time series pattern of the biometric information test data is similar.
  • a facility or the like that satisfies the above extraction conditions may be searched.
  • advice is given to go to that particular medical institution.
  • the control unit 1 determines in cooperation with the communication control unit 1a whether or not a database suitable for searching / inferring using the specific information determined in step S3 is stored in the DB unit 8. For example, as a result of the determination in step S3, as a result of inspecting the color of the stool, when the numerical value representing red becomes larger than that in the case of health, a database suitable for determining such a health condition is stored in the DB section. It is determined whether or not it is accumulated in 8. Even if a suitable database is not stored in the DB unit 8 in the information transmission system including the control unit 1, it may exist in another system. Therefore, the database may be searched including other systems.
  • step S5 when the control unit 1 determines that the database exists in cooperation with the communication control unit 1a, the control unit 1 cooperates with the communication control unit 1a and is recorded in the DB unit 8.
  • Acquire related facility information including equipment (S7) what kind of temporal changes and transitions are made in the data output by the above-mentioned information determination device 2 of a specific patient candidate (test data, biological data, vital data, sample data, etc. are generally expressed as "time series data").
  • Search for people with similar patterns from the database (DB section 8) based on logic such as pattern matching, search for people who are consulting at a specific medical institution, and extract the characteristics of that medical institution. do it.
  • the above-mentioned database acquires inspection data in chronological order using an inspection device having a specific specification, acquires change pattern information of the inspection data, and responds to this change pattern information by taking medication or improving life.
  • specific improvement timing information such as start as time information similar to the time information of the change pattern information, it has information that makes it possible to compare the part of the change pattern before the improvement information.
  • an information transmission method characterized by transmitting improvement information to a specific person by comparing a part of a change pattern of a specific person with a part of a change pattern before the improvement information of another person. ..
  • the above-mentioned database is created by acquiring a plurality of inspection data in time series using an inspection device having a specific specification and a specific inspection data change pattern information. Each timing of a plurality of event information regarding the acquisition source of the inspection data within the period is recorded in association with each other. Using this database, it is possible to extract and provide information on events that have influenced the trend change from the event information based on the relationship between the trend change of the change pattern and the timing of the event.
  • step S5 it becomes possible to acquire related facility information and lifestyle information suitable for examining and treating a specific disease suspected from the specific information acquired in step S3.
  • the related facility information possessing the equipment used for inspecting and treating a specific disease is acquired, and this facility information is used as a candidate.
  • the above-mentioned stool is red, hemorrhoids or colon cancer are suspected, but since more detailed classification is possible based on the characteristics of color and the characteristics of time-series changes, for example, the anus. It will be possible to sort out and obtain information about facilities that have colonoscopies and can perform colonoscopy instead of departments.
  • step S5 If there is no database as a result of the determination in step S5, inference is made using historical data (S9). It is possible that a database suitable for the specific information acquired in step S3 is not prepared.
  • the historical data is input to the inference engine 7, and the inference engine 7 uses the inference model to infer the facilities and the like capable of appropriate inspection. This infers diseases related to specific information, and infers clinical departments / departments related to specific information. Then, based on this inference result, an examination facility (institution) / medical care facility (institution) suitable for the user's health condition is recommended. In addition to inferring the recommended medical facility, this step also infers the name of the user's disease and the time when the user's symptoms worsen and go to the clinic for treatment. The detailed operation of inference using this historical data will be described later with reference to FIG.
  • the recommended facility is narrowed down based on the user's profile and the like (S11).
  • the user goes to the hospital based on the profile information acquired in step S15, the information acquired from the related inspection institution 9, for example, the clinic information near the user, and the hospital information on the user's holiday. You may choose a convenient facility. This selection is performed by the control unit 1 cooperating with the search unit 1f to acquire information suitable for the conditions from the information stored in the DB unit 8.
  • step S13 is a step of providing information on examinations and medical assistance to users and related persons, and is displayed on the terminal 4. Further, depending on the health condition of the user, a warning display may be displayed.
  • the control unit 1 acquires the sensor detection results from the information determination device 2 and the like (S1), and is related to the health state (disease) from these detection results. It is determined whether or not there is specific information to be used (S3). When there is specific information, the database is searched to search for facilities suitable for examining and diagnosing the health condition (disease), including the owned equipment (S7). For this reason, the user can perform health checks in daily life, and can receive advice on facilities suitable for examinations and medical examinations, considering the facilities owned by the facilities according to the health condition. ..
  • the DB search (S7) and the inference (S9) are treated independently as separate processes.
  • the present invention is not limited to this, and these may be treated comprehensively.
  • step S9 of FIG. 4 the operation of "inference using historical data" in step S9 of FIG. 4 will be described using the flowchart shown in FIG.
  • the control unit 1 performs the operation in cooperation with the inference engine 7, the DB unit 8, and the like through the communication control unit 1a.
  • time series data is acquired (S21).
  • the time-series data corresponding to the specific ID recorded in the DB unit 8 is acquired (see FIG. 3 for the time-series data).
  • the time width of the time-series data to be acquired is set to a specific time width, but if the data of the specific time width cannot be acquired, it is set to the time range that can be acquired. This is because if there is no specific time width, the judgment will be based only on the data under a specific situation, and the reliability will be inferior.
  • the specific time span differs between diseases that progress over time, such as colorectal cancer, and diseases that progress in a short period of time, such as influenza.
  • the type of test data depends on the learning of the inference model, but it is desirable that the graph is a specific item used at the time of learning. For example, it is preferable not to infer body weight and blood pressure together.
  • step S23 After acquiring the time series data, it is next determined whether or not the change is remarkable (S23). Here, it is determined whether or not the change pattern of the acquired time series data is remarkable. If the result of this determination is significant, a warning is given (S25). If the value of the specific information itself deviates significantly from the standard value, or if the change pattern is remarkable, there is a high possibility that there is some kind of disease, so a warning is displayed to the user without executing the process of step S27. To do. This warning display may be displayed on the terminal 4, for example. When a warning is displayed, this flow is terminated and the original flow is restored.
  • step S27 it is determined whether or not the level has no problem.
  • the change pattern of the time series data acquired in step S21 is at a level at which there is no problem.
  • the level of the inspection data is equal to or lower than the predetermined level when viewed in a specific time width, it is determined that there is no problem.
  • the medical institution information is not output (S29). In particular, since there is no information to be displayed to the user, information about the medical institution is not output.
  • step S29 is executed, the flow returns to the original flow.
  • step S27 it is determined whether or not the time series data for a specific time width can be acquired (S31). For example, if the status of occult blood is detected, it is determined whether or not occult blood was obtained within a range of several months. That is, the specific time span depends on the disease involved.
  • the data is input to the inference unit and the inference result is acquired (S35). Only when the data of a specific time width can be acquired, the data is input to the inference engine 7 and the inference result is obtained.
  • the learning unit 5 can generate an inference model that can output an inference result such as "many people go to a clinic with a XX inspection device" including recommended equipment.
  • teacher data based on time-series data, whether or not it is necessary to go to the hospital, when it is necessary to go to the hospital, and which clinical department to go to.
  • Various inference models can be generated, such as whether to do so.
  • step S35 since the database organized as shown in FIG. 2 is used (used for annotation of the biological data change pattern), the medical institution is not simply output as the inference result. , Various features of the facility are output as inference results, and results that do not depend on individual medical institutions can be obtained.
  • the biological data change pattern used for learning receives medical treatment by selecting and learning the data before it is improved by going to the hospital, taking medication, etc. when the data of the person who has already visited the hospital is used as the teacher data. It will be easier to compare with the biometric data patterns of those who are wondering whether or not they are, or those who have no subjective symptoms and are not receiving medical treatment.
  • the database according to the embodiment of the present application acquires time-series change pattern information using an inspection device having a specific specification, and similarly performs specific improvement timing information such as taking medication or starting life improvement with respect to the change pattern information. By including it as time information, it has information that makes it possible to convert the change pattern before the improvement information into teacher data. Further, in the information transmission method in the embodiment of the present application, the change pattern of a specific person is input to the change pattern before the improvement information of another person, and the improvement information obtained as an inference result is transmitted to the specific person. ing.
  • step S31 if the data of the specific time width has not been acquired, no inference is performed (S33). Inference is possible depending on the expected reliability without information of a specific time width, but it can be difficult. Therefore, if it is determined in step S31 that the data of the specific time width has not been acquired, no inference is made. However, there are cases where a clearly dangerous situation can be detected, in which case emergency information may be output before inference.
  • the acquired data is a numerical value that has a significant problem
  • there is no time grace to make a long-term forecast by inference and it will be dealt with in steps S23 and S25.
  • This response prevents the problem of not being able to respond to an emergency without making inferences in step S33, and makes it possible to create a highly reliable system in which information is output after sufficient data has been collected. That is, in the present embodiment, in the case of a numerical change that is contained in a specific change, the change pattern of the test data of the subject is cut out within a predetermined time width, and inference is performed according to the inference model learned together with the time information. When the inference result is obtained, this flow is terminated and the original flow is restored.
  • the level of the time series data is at a level that does not cause any problem, and if there is no problem, no information is output (S29).
  • the time series data is at a level where there is no problem, it is possible to refrain from outputting the information in order to prevent unnecessary inference. That is, when there is a disease having a cause other than the change in biological information, the possibility of outputting an unrelated inference result when this information is included in the teacher data is excluded. That is, in the present embodiment, when the specific criteria are not satisfied (for example, S23Yes, S27Yes), the inference model learned together with the time information by cutting out the change pattern of the inspection data of the subject within a predetermined time width. No inference is made according to. We expect the learning effect to be excellent in inferring something from subtle changes that cannot be visually detected by humans, and eliminate the harmful effects of making too subtle inferences.
  • the inference by the inference engine 7 is executed in step S35.
  • the determination in steps S23 and S27 may be made by the inference engine 7. In this case, it may be executed using the same inference model, or it may be executed by different inference engines.
  • step S7 of FIG. 4 the database was simply searched.
  • the acquired time-series data had to be related to a specific clinical department, but in reality, information not related to a specific clinical department may be required. For example, gaining weight or increasing blood pressure may require tests other than specific departments or specific devices.
  • related facility information can be appropriately acquired even in such a case.
  • control unit 1 When the control unit 1 starts the operation of the flow shown in FIG. 6 in cooperation with the communication control unit 1a, it first determines whether or not the information is for a specific clinical department (S41). Here, the control unit 1 determines whether or not the time-series data to be acquired is information for a specific clinical department.
  • step S41 if it is not for a specific clinical department, in steps S43 to S47, in addition to the data recorded in the DB unit 8, a plurality of information (data) transitions are used to be appropriate.
  • the control unit 1 extracts a change pattern of the inspection data of the subject within a predetermined time width, and extracts the data that changes with time.
  • An inference model is generated by learning using this inference model, inference is performed using this inference model, and information is output by referring to the inference result and the possessed device database.
  • the inference model used here includes data with time information (see Dy (t) in FIG. 2) as shown in FIGS. 3 (a) to 3 (c) during learning and the clinical department received by the user. It is assumed that it is learned as an output.
  • step S43 time series data is acquired (S43).
  • the control unit 1 acquires the past data from the information determination device 2 and the like as a history. These data are recorded in the memory in the control unit 1 or the DB unit 8.
  • the acquired time series data is data for a specific time width (S45).
  • a general clinic is recommended (S53). Since the time width of the time series data does not have a specific time, it is not possible to infer related diseases and refer to a specific facility, so a general medical facility is recommended. Note that the medical institution information may not be output as in S29 of FIG.
  • the data is input to the inference unit and the related disease information is acquired (S47).
  • the control unit 1 inputs the acquired data to the inference engine 7 and infers the related disease information.
  • the inference model used at this time generates information as shown in FIG. 3 as teacher data.
  • step S47 If the related disease information is acquired in step S47, or as a result of the determination in step S41, the information is for a specific clinical department, then the related equipment is determined (S49), and the equipment possession facility information (clinical department) is determined. ) Is acquired (S51).
  • the control unit 1 determines what the specific clinical department determined in step S41 or the equipment related to the disease acquired in step S47 is. Then, the control unit 1 searches for and selects a facility having the determined equipment from the facility-specific equipment list 8a of the DB unit 8.
  • each ID is determined based on the sensor output result (S1).
  • the control unit 1 determines the inspection result for each ID attached to the sensor output result based on the sensor output.
  • the specific information related to the disease for example, a feature such as a numerical value different from the healthy state, based on the determination in step S1. judge. If the specific information cannot be obtained as a result of the determination in step S3, the user's profile, behavior, and lifestyle are determined in the same manner as in the flow of FIG. 4 (S15).
  • the process returns to step S1.
  • step S4 If specific information can be obtained as a result of the determination in step S3, it is determined whether or not there is an inference model (S4).
  • the inference model specification determination unit 1d in the control unit 1 determines the inference model specifications for giving advice (display of recommended facilities, etc.) to the target person, and the inference request unit 1e generates an inference model for the learning unit 5. To ask.
  • the inference model generated by the learning unit 5 is received, it is set in the inference model 7.
  • a plurality of inference models are created so as to correspond to various situations, and are stored in the control unit 1 or the inference engine 7.
  • this step 4 whether or not an inference model capable of giving appropriate advice (display of recommended facilities, etc.) to the target person for the specific information acquired in step S3 is stored in the control unit 1 or the like. To judge.
  • step S4 If there is an inference model as a result of the determination in step S4, inference is performed using the historical data (S6).
  • the inference model selected in step S4 is set in the inference engine 7.
  • the historical data of the target person (see the time-series data in FIG. 2) transmitted from the information determination device 2 and recorded in the DB 8 is input to the input layer of the inference engine 7, and inference is performed.
  • the inference result is output from the output layer.
  • the inference in step 6 may be performed by executing the flow shown in FIG.
  • the inference result is acquired in step 6, it is next determined whether or not the inference result is highly reliable (S8).
  • the reliability of the inference result in step S6 is calculated, and it is determined whether or not this reliability is higher than the predetermined value.
  • the high reliability may be inferred using data prepared in advance, and the degree of deviation from this result may be calculated.
  • step S10 If the result of the determination in step S8 is low reliability, or if the result of the determination in step S4 is that there is no inference model, training data is collected and an inference model is created (S10).
  • the control unit 1 collects the specific information and similar data acquired in step S3.
  • various databases such as a large number of DB units 8 connected on the Internet and databases in related inspection institutions 9 and the like are searched, and the above-mentioned similar data is collected.
  • the data shows the name of the disease, test items, and the like.
  • time-series data that shows what kind of biological information it is (or what kind of specifications and properties it is measured by measuring devices, sensors, etc.), and also has signs and precursors of assumed diseases. It is preferable that the number of data and the acquisition time range are sufficient to read the time-series pattern that indicates that the patient has a disease.
  • teacher data is created by associating the output result with the input data for each similar data.
  • annotation information for this pattern information such as whether or not there is information on the medical institution that led to the medical examination (more preferable if it includes information such as the clinical department and the equipment and fixtures there). I'm assuming. Teacher data generated using these patterns and annotation information is collected.
  • the control unit 1 requests the learning unit 5 to create an inference model through the inference request unit 1e.
  • the inference model learned from this teacher data is highly reliable and can be used to determine whether a person with that data pattern can maintain good health, see a doctor, and so on.
  • the reliability of the existing inference model is low, a new inference model is recreated from the teacher data, so that highly reliable inference can always be performed.
  • the learning unit 5 creates an inference model, it sends it to the control unit 1.
  • the control unit 1 receives the inference model, it executes steps S4 to S8 using this inference model.
  • the inference model for inferring the time when the user goes to the medical facility for medical examination, etc. In this user's state, it may end up going to the medical facility, but to avoid going. You may generate an inference model that advises you on how to deal with it. For example, if there is a description in the SNS or blog on the Internet that you are likely to get sick but have been treated, you may collect this information. In addition, if the medical database or the like contains information such as treatment for a person who is likely to get sick, such information may be collected. If information can be collected, learning is performed using this information and an inference model is generated.
  • the related inspection organization 9 shown in FIG. 1 is based on a technique of deriving and presenting information that contributes to the recovery and maintenance of the user's health by selecting and describing health-related information classified into a specific category.
  • the writing and selection information input by the person should be annotated together with the historical data of the specific target person (see the time series data in FIG. 2) as the teacher data.
  • the teacher data For example, it is possible to convert the inference model into teacher data.
  • doctors and medical professionals may also use the services described above to refer to the diagnosis, so it is used by doctors at the time of medical examination / diagnosis performed by doctors at the time of medical examination of a specific patient.
  • the manual input result to an information terminal such as a PC or the voice input result may be used.
  • the patient's history data (see the time-series data in FIG. 2) is acquired, and the above-mentioned input result is added to this history data as annotation information so that which item corresponds to what kind of input information can be identified. I will go.
  • the patient history data does not necessarily have to be the history data of one item, and may be used by incorporating information on lifestyle habits, hospital visits, medication information, etc. in chronological order.
  • Using the teacher data obtained in this way we train to obtain reliable results and create a new inference model.
  • inspection data is acquired in time series using an inspection device having a specific specification
  • change pattern information of the inspection data is acquired
  • a specific health advice service is provided for this change pattern information.
  • the inference model in which the result of annotating the input information of is trained as teacher data, the result of inferring by inputting the inspection data obtained in time series using the inspection equipment of the specific specifications of a specific person is specified. It becomes possible to provide a method of transmitting information to the person in question.
  • this information transmission method first responds to the input of a specific health response event, and prior to the health response event, uses a specific specification inspection device to respond to the specific health response event in chronological order.
  • the inspection data is acquired and the time-series inspection data is recorded.
  • the result of annotating the change pattern information of the recorded test data with the information of the equipment, equipment, and environment at the time of the health response event is created as teacher data.
  • learning is performed and an inference model is generated.
  • inspection data is acquired in chronological order using an inspection device with specific specifications, and the acquired inspection data is input to an inference model to infer information on equipment, equipment, and the environment. From this inference result, a health response event customized for a specific person is transmitted to a specific person.
  • the health response event is an event related to health, for example, an event such as receiving a medical examination by a doctor at a medical facility.
  • step S8 If the result of the determination in step S8 is high reliability, then the recommended facility is displayed based on the inference result (S13a).
  • the recommended facility is displayed on the terminal 4 based on the inference result in step S6.
  • the medical facilities and inspection institutions that can carry out this inspection may be searched for the facility-specific possessed equipment list 8a in the DB8, and the search results may be displayed. Further, in step S6, if it is possible to infer a medical facility or an institution that can be inspected, the recommended facility may be displayed based on the inference result.
  • step S13a advice on recommended facilities, etc. is obtained using the biometric information test data for each individual obtained in chronological order. Furthermore, the result of annotating the characteristic information of medical institutions, etc. that had an improvement effect on the change pattern information of the biometric information test data, or the result of annotating the input information of a specific health advice service is learned as teacher data. The method of generating the inference model and using this inference model to obtain the trigger information when improving the inspection data is explained.
  • the information that triggers the improvement of the test data obtained here is information that has been reduced to the level of the test or treatment, the life pattern and life of the person who transmitted the information (the person who receives the information). It is desirable to devise ways to convert and present advice so that information that can be used is transmitted even in the regions and regions where it is used. By searching for information that meets the conditions in light of the lives of the people who receive the information, it will be possible to obtain information that can be used in other regions as well. In addition to the living area of the person who receives the information or the area adjacent to it, it is sufficient to search for facilities that satisfy the extraction conditions and convey the corresponding items by display or the like.
  • step S13a When the recommended facility is displayed in step S13a, the process returns to step S1.
  • inference is performed using an inference model in various steps (for example, S9 in FIG. 4, S35 in FIG. 5, and S6 in FIG. 7).
  • the inference model can also input patterns of time-series changes in biometric information to infer future health risks and their remedies, or to infer how far in the future it is.
  • the creation of such an inference model will be specifically described with reference to FIGS. 8A and 8B.
  • FIG. 8A shows the temporal change of the health state
  • the horizontal axis shows the change of time T
  • the vertical axis shows the biological data D.
  • the time-series data Dts11 and Dts12 in which the black circles in FIG. 8A are connected indicate the case where the health condition deteriorates due to some trigger (specific event), and the time-series data Dts21 and Dts22 in which the white circles are connected are the triggers (specification).
  • the event shows when the health condition is improved.
  • the health condition is improved or deteriorated by some trigger (specific event; timing is indicated by the trigger information Inf), the health condition appears in the biological data D, and the trend of the pattern changes.
  • the vertical line shown by the dotted line parallel to the trigger information Inf is an example of another event.
  • an increase in biological data D indicates a deviation from the health state
  • a decrease in biological data D indicates recovery or improvement. For example, eating something bad (this is an event) and having a fever, or having a medical examination and taking medicine (this is an event) and recovering can be familiar, but in many cases what is the trigger. There are few patients who do not know or are aware of how they have recovered numerically.
  • time-series data Dts21 and Dts22 are different numerical values, and those that have decreased due to changes in health status (for example, muscle strength, bone density, and visual acuity due to aging) (Dts21) are events such as going to a hospital and receiving treatment. (Inf timing) may be taken as an example showing how it has recovered.
  • FIG. 8B By executing such a flowchart, it is possible to identify an event of what affects health and how.
  • the flow shown in FIG. 8B is executed by controlling each unit in the control unit 1 by a CPU or the like in the control unit 1, but is not limited to the control unit 1 and may be executed by another server, engine, or the like. ..
  • the event information corresponding to the time series data acquisition period is acquired (S61).
  • the time-series data acquisition period varies depending on the target disease, health condition, etc., but here, it may be determined in consideration of the period in which the event generally affects the health.
  • the event information is acquired from various sources such as a schedule recorded on the user's terminal 4 and a schedule uploaded to SNS or the like.
  • inspection data is acquired in a specific time series using an inspection device having specific specifications.
  • the event was attended by the subject, such as going to a hospital for medical examination, purchasing medicine at a pharmacy and taking it, or going to a cold and crowded place. Actions and behaviors that may affect your health.
  • step S63 the trend change of the pattern before and after the specific event is compared, and it is determined whether or not there is a difference (S63).
  • the specific event information is acquired in step S61, it is determined whether or not there is a change in the trend of the time series data (biological data) before and after the specific event. That is, the change pattern information of the inspection data is acquired, and it is determined whether or not the pattern of the time series data before the specific event and the trend (trend) of the time series data after the specific event are the same.
  • step S63 If there is no difference in the trend change as a result of the determination in step S63, the determination is made using another event (S75). In this case, since there is no difference in the trend change before and after the event in step S63, it can be said that the health state does not change before and after the event. Therefore, the process returns to step S63, and an event with a trend change is searched for.
  • step S65 the trend of the data change before and after the trigger information (Inf in FIG. 8A) is classified (S65).
  • Trigger information You can see whether the event affected your health when there is no difference in the trend change before and after. However, since it may have a good effect or a bad effect, in this step S65, it is determined whether the trend is a good trend or a bad trend, and the trend of the data change pattern is classified based on the determination result.
  • step S65 the classification result in step S65 is classified into an improved event or a worsened event, and the environment, components, and the like of the event are used as annotation information.
  • the environment and components are the decomposition of an event, such as simply going to a crowded place on a cold day, instead of simply going to Tokyo Station.
  • the event "I went to the hospital and had a medical examination” was simply held at an event such as "I took an X-ray", “I injected XX", "I took ⁇ ", etc. It is a disassembled version of. That is, in step S67, the event information is decomposed for each configuration and the components are extracted. By extracting the components and the like, creating teacher data, and learning, it is possible to provide customized information based on the extracted components that have affected the trend change.
  • the pattern showing the trend change before or after the specific event is converted into teacher data (S69).
  • the good event and the bad event are annotated with respect to the time-series information pattern indicating the trend change before the specific event and converted into teacher data.
  • step S71 the learning / inference model is created (S71).
  • learning is performed using the teacher data created in step S69 to generate an inference model. That is, when the trend changes, there is an event that triggers it, so when time series data is input, an inference model that can output a specific event (impact event) is generated.
  • step S69 if the teacher data is generated using the pattern before the specific event, it is possible to generate an inference model capable of inferring that the presence of the specific event makes it worse (in this way).
  • teacher data is generated using the pattern after a specific event, it is possible to generate an inference model that can make an inference that if there is the specific event, it will be improved (in this way).
  • this step S71 with respect to the change pattern information, the characteristic information of the medical institution or the like having the improvement effect and the influence event information at the time corresponding to the time when the conversion pattern changes within a specific period are detected and detected.
  • an inference model When an inference model is generated, it is updated at a specific timing (S73). Biological data and event information are accumulated from moment to moment, so they are updated regularly or at the time of the weather, news, etc. If a specific data pattern is input to the inference model generated in step S71, it is possible to give advice that can be improved or deteriorated in this event in the future, so that it is necessary to be aware of it. For seasonal changes and diseases that are prone to infection, it is better to use information such as moving to congested areas as an event. This may be added in step S73 so as to correspond to the time axis of the time series data. You can also create an inference model that determines what went wrong with data changes after the event.
  • the flow of creating the inference model is a step of determining an event that affects the change pattern of the time-series data of the biological information representing the health state (S63) and a step of determining whether the event has an improvement effect (S63).
  • An information providing method including S67) and a step (S69) of determining an event having an effect of improving a change pattern according to a time-series pattern before the time of the event is shown. That is, it is possible to discover events that affect health based on the post-event pattern.
  • the recording system in the present embodiment acquires a plurality of inspection data in chronological order over a specific period using a (biological data) inspection device having a specific specification, and creates a data change pattern.
  • a plurality of event information regarding the data acquisition source within this specific period can be recorded in association with each other up to the timing of when it occurred.
  • This recording system may include a database for recording data and events. Further, this database may be a plurality of linked computers. Since this recording system should be able to extract the events that affected the trend change, it provides useful information to other people by making it a circuit, program, or system that can provide information on the events that have affected this trend. sell.
  • the event information is subdivided based on the environment and situation where the event occurred and other component information obtained, it can be provided as customized information replaced by the position of the person who provides the information. For example, if you break it down into factors such as going to a crowd in the area where the outbreak of an infectious disease was reported, it can be generalized from the actual specific behavioral information that you went to the ferry landing, so don't go to a crowded place today. It is converted into information such as.
  • the inspection data change pattern and the teacher data of many cases obtained by annotation of these influence events are collected, learned and an inference model is created, the change pattern is input to the inference model, and the output is the influence event element. You may try to.
  • inspection data is acquired in a specific time series using an inspection device having specific specifications (see, for example, S61 in FIG. 8B).
  • the change pattern information of the test data was acquired (for example, see S63 in FIG. 8 (b)), and the characteristic information of the medical institution etc. that had an improvement effect on the change pattern information and the conversion pattern changed within a specific period.
  • the inspection data obtained in time series in a period similar to the width of the specific period is input to the inference model to infer the influence event, and the inference is performed.
  • the results are transmitted (for example, in S9 in FIG. 4, S35 in FIG. 5, and S6 in FIG. 7, it can be inferred using the above-mentioned inference model).
  • inspection data is acquired and recorded in time series using an inspection device having a specific specification prior to the health response event.
  • An inference model is acquired in which the result of annotating the equipment, equipment, and environment information at the time of the health response event is learned as teacher data for the change pattern information of the test data (for example,). See S71 in FIG. 8 (b)).
  • the inspection data obtained in time series using the inspection device of the specific specification of the specific person is input to the above-mentioned inference model and inferred, and the inference result is transmitted to the specific person (for example).
  • S9 in FIG. 4, S35 in FIG. 5, and S6 in FIG. 7 can be inferred using the above-mentioned inference model).
  • inspection data is acquired in time series using an inspection device having specific specifications, and change pattern information of the inspection data is acquired (see, for example, S1 in FIG. 4).
  • change pattern information such as taking medication and starting life improvement as the same time information for the change pattern information
  • a database having information that makes it possible to compare the change patterns before the improvement information is created.
  • the change pattern of a specific person is compared with the change pattern of another person before the improvement information, the improvement information is searched, and the searched improvement information is transmitted to a specific person (for example).
  • information can be transmitted using the search results of the database).
  • change pattern information of inspection data created by acquiring a plurality of inspection data in a time series using an inspection device having a specific specification and a specific inspection data.
  • event information it is possible to provide information on events that have influenced the trend change (see, for example, FIGS. 8A and 8B).
  • a database may be searched, or an inference model may be generated and inferred using this inference model.
  • various events such as impact events and health response events are described, but these are also included in the broad sense of the event.
  • the inspection data of the subject is acquired (for example, see S1 in FIG. 7), and it is determined whether or not the inspection data is specific input data (for example, in FIG. 7). (See S3), and if the inspection data is specific input data as a result of this determination, inference is performed using a specific inference model (see, for example, S4Yes ⁇ S6 in FIG. 7), and the specific input data is subjected to inference.
  • a specific inference model see, for example, S4Yes ⁇ S6 in FIG. 7
  • training is performed to generate an inference model by newly collecting input data similar to the specific input data and using it as teacher data (for example).
  • the inspection data of the subject is input (for example, refer to the information determination device 2 in FIG. 1, S1 in FIG. 4, S21 in FIG. 5 and the like), and the profile information of the subject and the profile information of the subject.
  • the possessed device information for each test / medical institution see, for example, DB section 8 in FIG. 1, S7 in FIG. 4 and the like
  • the transmission information to be transmitted to the target person is determined (see, for example, the information providing unit 1c in FIG. 1, S11 in FIG. 4 and the like).
  • the information can be transmitted to the target person in consideration of the equipment owned by the examination / medical institution. Further, in the present embodiment, it is possible to introduce an effective facility based on the health-related information obtained unconsciously on a daily basis, and it is possible to save the trouble of visiting the facility many times.
  • the test data of the subject is input (see, for example, the information determination device 2 in FIG. 1, S21 in FIG. 5 and the like), and the test of the subject / visit to a medical institution / inspection.
  • the inference unit that has an inference model learned according to the medication information
  • the change pattern of the test data of the subject is input and inference is performed, and the transmission information to be transmitted to the subject is determined based on the inference result.
  • the information providing unit 1c of FIG. 1, the inference engine 7, FIG. 3 (a), S35 of FIG. 5 and the like Therefore, it is possible to notify the subject by predicting the time of visit to a medical institution or the like in advance from the time-series change of the examination data. It will be possible to receive appropriate tests and treatment before the symptoms worsen.
  • the information transmission program includes a step of inputting test data of a subject (see, for example, information determination device 2 in FIG. 1, S21 in FIG. 5 and the like), and inspection / medical treatment of the subject.
  • test results and test results are stored in a database and the test results are used, there is no description about notifying the optimal medical institution based on the test results.
  • proposals to ask medical personnel in remote areas to observe excrement there is no proposal to inform the optimal medical institution.
  • proposals to display a history browser there is no proposal to announce the optimal facility for examining the symptoms of a subject based on the equipment owned by a medical institution or the like.
  • Many general users unknowingly install sensors, surveillance cameras, watching cameras, home thermometers, weight scales, and bodies in sanitary facilities such as mobile terminals, home appliances, baths, toilets, and washrooms in their daily lives.
  • Biological information may be monitored in chronological order with a composition meter, sphygmomanometer, or the like.
  • daily monitoring is performed, information obtained by the monitoring is actively utilized, and advice is provided so that a detailed examination can be performed at an appropriate facility based on biological information and sample information.
  • the information determination device 2 may be any device for acquiring health-related information of the target person, for example, vital information, sample information, and the like. In the simplest example, it can be applied to face image information obtained from a mobile terminal such as a smartphone, heartbeat information based on the face image information, and the like, and these information may be utilized. In addition, it may be linked with a device such as a wearable terminal that is used in close contact with the user, and data to be noted such as arrhythmia can be easily acquired by these devices.
  • a history pattern that includes multiple data instead of analyzing single-shot data that may contain errors depending on the equipment, physical condition, eating and drinking, and living scenes, the presence or absence of illness, possibility, recovery, and outpatient visits Information such as when to do it, advice information, etc. can be provided with high accuracy. If this information is of low accuracy, the user will be delayed in seeing the doctor and will be more concerned than necessary.
  • control unit 1 has been described as an IT device composed of a CPU, a memory, an HDD, and the like.
  • part or all of each part may be configured in a hardware circuit, and a gate generated based on the program language described by Verilog.
  • a hardware configuration such as a circuit may be used, or a hardware configuration using software such as a DSP (Digital Signal Processor) may be used.
  • DSP Digital Signal Processor
  • the element is not limited to the CPU, and may be any element that functions as a controller, and the processing of each part described above may be performed by one or more processors configured as hardware.
  • each part may be a processor each of which is configured as an electronic circuit, or may be each circuit part of a processor composed of an integrated circuit such as an FPGA (Field Programmable Gate Array).
  • a processor composed of one or more CPUs may execute the functions of each unit by reading and executing the computer program recorded on the recording medium.
  • the controls mainly described in the flowchart can often be set by a program, and may be stored in a recording medium or a recording unit.
  • the recording method to the recording medium and the recording unit may be recorded at the time of product shipment, the distributed recording medium may be used, or may be downloaded via the Internet.
  • the operation in the present embodiment has been described using a flowchart, but the order of the processing procedures may be changed, or any step may be omitted. Steps may be added, and specific processing contents in each step may be changed.
  • the present invention is not limited to the above embodiment as it is, and at the implementation stage, the components can be modified and embodied within a range that does not deviate from the gist thereof.
  • various inventions can be formed by an appropriate combination of the plurality of components disclosed in the above-described embodiment. For example, some components of all the components shown in the embodiment may be deleted. In addition, components across different embodiments may be combined as appropriate.

Abstract

The objective of the present invention is to present an event relating to a change in the state of health of a user. According to the present invention: examination data are acquired as a time series in a specified time period, using an examination device having a specified specification (S61); variation pattern information relating to the examination data are acquired (S63); influencing event information at a time point corresponding to the time point at which the variation pattern varied within the specified time period is detected with respect to the variation pattern information, and an inference model that has been taught using, as teacher data, results obtained by annotating the detected influencing event information, is acquired (S65 to S71); an influencing event is inferred by inputting into the inference model examination data of a specified person, obtained as a time series in a time period having a similar width to the specified time period, using the inspecting machine having the specified specification; and the influence result is transmitted.

Description

情報伝達装置および情報伝達方法Information transmission device and information transmission method
　本発明は、日常生活において取得可能な検査結果に応じてアドバイス等のカスタマイズ情報をユーザに提供することができる情報伝達装置および情報伝達方法に関する。 The present invention relates to an information transmission device and an information transmission method capable of providing a user with customized information such as advice according to an inspection result that can be obtained in daily life.
　近年インターネットが普及してきており、インターネットを利用することによって、ユーザの生活に密着した情報を容易に取得することができる。この取得した情報を利用することによって、個々のユーザに相応しい様々なカスタマイズ情報（有効情報）を生成し、このカスタマイズ情報を提供するサービスが増えてきている。例えば、健康食品などを紹介するサービスは、多くの人に共通する興味を引く情報であることから、この種のサービスが多く見受けられる。 The Internet has become widespread in recent years, and by using the Internet, it is possible to easily obtain information closely related to the user's life. By using this acquired information, various customized information (valid information) suitable for each user is generated, and services that provide this customized information are increasing. For example, services that introduce health foods are often interesting information that is common to many people, so this type of service is often found.
　また、ネットワーク環境が整備されてきていることから、病院などの専門機関の外部において、遠隔検査を行うことが種々提案されている。例えば、特許文献１には、センサチップと携帯電話をリーダ／ライタとして用い、公共の通信網を利用することによって、検査データを送信する遠隔検査方法が開示されている。そして、この特許文献１には、過去の検査データやその評価結果をデータベースに蓄積し、これらの情報を利用することが提案されている。 In addition, since the network environment has been improved, various proposals have been made to perform remote inspections outside specialized institutions such as hospitals. For example, Patent Document 1 discloses a remote inspection method for transmitting inspection data by using a sensor chip and a mobile phone as a reader / writer and using a public communication network. Then, in Patent Document 1, it is proposed to store past inspection data and its evaluation results in a database and use these information.
　また、特許文献２には、個人認証データと、排泄物撮影手段によって撮影された画像データと、を統合し、この統合したデータを通信手段によって送信する生体情報測定装置が開示されている。さらに、特許文献３には、検査に関するスタディリストを表示し、選択されたスタディにおける被検者の医療画像情報による医療画像を画像表示画面に表示し、ヒストリーブラウザを要求すると、被検者のスタディのリストを表すヒストリーブラウザを表示する医療情報の表示方法が開示されている。 Further, Patent Document 2 discloses a biometric information measuring device that integrates personal authentication data and image data photographed by excrement photographing means and transmits the integrated data by communication means. Further, in Patent Document 3, a study list related to an examination is displayed, a medical image based on the medical image information of the subject in the selected study is displayed on the image display screen, and when a history browser is requested, the study of the subject is requested. A method of displaying medical information that displays a history browser representing a list of medical information is disclosed.
　また、特許文献４には、患者の状況や症状に応じた情報を、患者の問診情報、検査値等の過去の診断情報、病院情報等に基づいて、提示することが開示されている。特許文献５には、医療担当者が患者の時系列の検査データの所定時間軸での変化パターンをチェックすることが開示されている。特許文献６、特許文献７および特許文献８には、時系列データを推論エンジンに入力して推論結果を得ることが開示されている。特許文献９には最適な検査間隔の決定を支援することが開示されている。特許文献１０には、血液、尿、大便等の検体の生化学検査の結果データを用いてモデル学習を行い、このモデルを用いて情報を提供することが開示されている。 Further, Patent Document 4 discloses that information according to the patient's situation and symptoms is presented based on the patient's interview information, past diagnostic information such as test values, hospital information, and the like. Patent Document 5 discloses that a medical person checks a change pattern of a patient's time-series examination data on a predetermined time axis. Patent Document 6, Patent Document 7, and Patent Document 8 disclose that time series data is input to an inference engine to obtain an inference result. Patent Document 9 discloses that it assists in determining the optimum inspection interval. Patent Document 10 discloses that model learning is performed using the result data of biochemical tests of samples such as blood, urine, and stool, and information is provided using this model.
特開２００９－２５８８８６号公報JP-A-2009-258886 特開２０１４－０３１６５５号公報Japanese Unexamined Patent Publication No. 2014-031655 特許第５２９４９４７号公報Japanese Patent No. 5294947 特開２０１６－０１８４５７号公報Japanese Unexamined Patent Publication No. 2016-018457 特開平１１－０８９８２２号公報Japanese Unexamined Patent Publication No. 11-089822 特開２０１８－５１８２０７号公報JP-A-2018-518207 国際公開２０１９／０２２０８５号公報International Publication No. 2019/022085 特開２００６－５１１８８１号公報Japanese Unexamined Patent Publication No. 2006-511881 特開２０１１－２５３４６４号公報Japanese Unexamined Patent Publication No. 2011-253464 特開２０１９－２１１８６６号公報Japanese Unexamined Patent Publication No. 2019-21186
　前述した特許文献１－３には、生体情報を取得し、この情報を遠隔に送信し利用することが記載されている。被検者が自覚症状を有さない場合には、病気にかかっていることや病気になる可能性を知ることができ有益である。しかし、特許文献１－３には、個人のプロフィールに従って、アドバイス等のカスタマイズ情報を生成し、このカスタマイズ情報をユーザに知らせることについて何ら記載されていない。 The above-mentioned Patent Document 1-3 describes that biometric information is acquired and this information is remotely transmitted and used. If the subject has no subjective symptoms, it is useful to know that he / she is ill and that he / she may become ill. However, Patent Document 1-3 does not describe anything about generating customized information such as advice according to an individual's profile and notifying the user of this customized information.
　特許文献６－８には、推論モデルを用いて推論を行い、被検者に情報を提供することが開示されている。しかし、検査データは、あるイベント（きっかけ）があると、大きく変化する場合がある。このようなユーザの健康状態に影響を与える影響イベントを提示することについては、何ら記載されていない。ユーザの健康状態等が変化した際に影響を与えた影響イベントが分かると、以後、ユーザは影響イベントに注目して健康を維持することが可能となる。 Patent Document 6-8 discloses that inference is performed using an inference model and information is provided to the subject. However, the inspection data may change significantly when there is a certain event (trigger). There is no mention of presenting such impact events that affect a user's health. Once the impact event that affected the change in the user's health condition is known, the user can subsequently pay attention to the impact event and maintain his / her health.
　本発明は、このような事情を鑑みてなされたものであり、ユーザの健康状態の変化に関連するイベントを提示できるようにした情報伝達装置および情報伝達方法を提供することを目的とする。 The present invention has been made in view of such circumstances, and an object of the present invention is to provide an information transmission device and an information transmission method capable of presenting an event related to a change in a user's health condition.
　上記目的を達成するため第１の発明に係る情報伝達方法は、特定の仕様の検査機器を用いて特定期間の時系列的に検査データを取得し、該検査データの変化パターン情報を取得し、上記変化パターン情報に対し、上記特定期間内で上記変化パターンが変化した時点に対応する時点における影響イベント情報を検出し、上記検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得し、特定の人物の上記特定の仕様の検査機器を用いて上記特定期間の幅と類似の期間で時系列的に得た検査データを、上記推論モデルに入力して影響イベントを推論し、該推論の結果を伝達する。　 In order to achieve the above object, the information transmission method according to the first invention acquires inspection data in a time series of a specific period using an inspection device having a specific specification, acquires change pattern information of the inspection data, and obtains the change pattern information of the inspection data. Inference that the effect event information at the time corresponding to the time when the change pattern changes within the specific period is detected for the change pattern information, and the result of annotating the detected effect event information is learned as teacher data. An impact event is inferred by acquiring a model and inputting inspection data obtained in time series in a period similar to the width of the specific period using the inspection equipment of the specific specifications of a specific person into the inference model. And convey the result of the inference.
　第２の発明に係る情報伝達方法は、特定の健康対応イベントの入力に対応し、上記健康対応イベントに先立って特定の仕様の検査機器を用いて時系列的に検査データを取得し記録されていた、該検査データの変化パターン情報に対し、上記健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報をアノテーションした結果を教師データとして学習させた推論モデルを取得し、特定の人物の上記特定の仕様の検査機器を用いて時系列的に得た検査データを、上記推論モデルに入力して設備、および／または備品、および／または環境の情報を推論し、該推論結果から上記特定の人物用にカスタマイズした健康対応イベントを、上記特定の人物に伝達する。 The information transmission method according to the second invention corresponds to the input of a specific health response event, and prior to the above health response event, inspection data is acquired and recorded in time series using an inspection device having specific specifications. In addition, an inference model is acquired in which the result of annotating the equipment and / or equipment and / or environment information at the time of the above-mentioned health response event is trained as teacher data with respect to the change pattern information of the test data. Then, the inspection data obtained in time series using the inspection equipment of the specific specifications of the specific person is input to the inference model to infer the information of the equipment and / or the equipment and / or the environment. From the inference result, the health response event customized for the specific person is transmitted to the specific person.
　第３の発明に係る情報伝達方法は、特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得して作成した上記データの変化パターン情報と、上記特定の期間内における上記検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録可能なデータベースを作成し、上記データベースを用いて、上記変化パターンのトレンド変化と上記イベントのタイミングの関係に基づいて、上記イベント情報のうち、上記トレンド変化に影響したイベントを抽出して情報提供を可能とする。 The information transmission method according to the third invention includes change pattern information of the above data created by acquiring a plurality of inspection data in a time series over a specific period using an inspection device having a specific specification, and the above-mentioned identification. Create a database that can record the timing of each of the plurality of event information related to the acquisition source of the inspection data within the period of the above, and use the database to determine the trend change of the change pattern and the timing of the event. Based on the relationship, it is possible to extract the events that have influenced the above-mentioned trend change from the above-mentioned event information and provide the information.
　第４の発明に係る情報伝達方法は、上記第３の発明において、上記イベント情報を構成毎に分解して構成要素を抽出し、上記トレンド変化に影響したイベント情報を抽出した上記構成要素に基づいてカスタマイズして情報提供する。 The information transmission method according to the fourth invention is based on the above-mentioned component in the above-mentioned third invention, in which the above-mentioned event information is decomposed for each component and components are extracted, and the event information affecting the above-mentioned trend change is extracted. Customize and provide information.
　第５の発明に係る情報伝達方法は、上記第４の発明において、上記データベースを用いて、上記トレンド変化に影響したイベントを提供するための　推論モデルは、検査データを取得し、該検査データの変化パターンを学習用推論部の入力とし、出力すべきアドバイスをアノテーション情報として、学習することによって推論モデルを生成し、この生成した推論モデルを用い、上記対象者の変化パターン情報を上記推論モデルに入力することにより推論結果を得て、この得られた推論結果に基づいて、上記伝達情報を決定する。 The information transmission method according to the fifth invention is the inference model for providing an event influencing the trend change by using the database in the fourth invention. The inference model acquires inspection data and obtains the inspection data. A change pattern is used as an input of the inference unit for learning, and advice to be output is used as annotation information to generate an inference model. Using this generated inference model, the change pattern information of the target person is used as the inference model. The inference result is obtained by inputting, and the above-mentioned transmission information is determined based on the obtained inference result.
　第６の発明に係る情報伝達装置は、特定の仕様の検査機器を用いて特定期間の時系列的に検査データを取得し、該検査データの変化パターン情報を取得するデータ取得部と、上記変化パターン情報に対し、上記特定期間内で上記変化パターンが変化した時点に対応する時点における影響イベント情報を検出し、上記検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する学習部と、特定の人物の上記特定の仕様の検査機器を用いて上記特定期間の幅と類似の期間で時系列的に得た検査データを、上記推論モデルに入力して影響イベントを推論する推論部と、上記推論の結果を伝達する情報伝達部と、を有する。 The information transmission device according to the sixth invention includes a data acquisition unit that acquires inspection data in a time-series manner for a specific period using an inspection device having a specific specification, and acquires change pattern information of the inspection data, and the above-mentioned change. For the pattern information, an inference model is obtained in which the influence event information at the time corresponding to the time when the change pattern changes within the specific period is detected and the result of annotating the detected influence event information is trained as teacher data. The influence event is input by inputting the inspection data obtained in time series in the width of the specific period and the period similar to the width of the specific period using the learning unit to be acquired and the inspection device of the specific specification of the specific person into the inference model. It has an inference unit for inferring and an information transmission unit for transmitting the result of the inference.
　第７の発明に係る情報伝達装置は、特定の健康対応イベントの入力に対応し、上記健康対応イベントに先立って特定の仕様の検査機器を用いて時系列的に検査データを取得し記録するデータ取得部と、記録されていた上記検査データについての変化パターン情報に対し、上記健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する学習部と、特定の人物の上記特定の仕様の検査機器を用いて時系列的に得た検査データを、上記推論モデルに入力して設備、および／または備品、および／または環境の情報を推論する推論部と、上記推論部による推論結果から上記特定の人物用にカスタマイズした健康対応イベントを、上記特定の人物に伝達する情報伝達部と、を有する。 The information transmission device according to the seventh invention corresponds to the input of a specific health response event, and data for acquiring and recording inspection data in a time series using an inspection device having specific specifications prior to the health response event. The result of annotating the information on the equipment, / or equipment, and / or environment at the time of the above-mentioned health response event with respect to the change pattern information about the above-mentioned inspection data recorded by the acquisition unit is used as the teacher data. Equipment and / or equipment by inputting the inspection data obtained in time series using the learning unit that acquires the trained inference model and the inspection equipment of the above-mentioned specific specifications of a specific person into the inference model. It also has an inference unit that infers environmental information and / or an information transmission unit that transmits a health response event customized for the specific person from the inference result by the inference unit to the specific person.
　第８の発明に係る情報伝達装置は、特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得するデータ取得部と、上記検査データを取得して作成した上記検査データの変化パターン情報と、上記特定の期間内における上記検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録可能なデータベースを作成するデータ作成部と、上記データベースを用いて、上記変化パターンのトレンド変化と上記イベントのタイミングの関係に基づいて、上記イベント情報のうち、上記トレンド変化に影響したイベントを抽出して情報提供を可能とする情報提供部と、を有する。 The information transmission device according to the eighth invention is created by a data acquisition unit that acquires a plurality of inspection data in a time-series manner over a specific period using an inspection device having a specific specification, and an inspection data that is acquired and created. A data creation unit that creates a database that can record the change pattern information of the inspection data and the timing of each of a plurality of event information related to the acquisition source of the inspection data within the specific period, and the database. Based on the relationship between the trend change of the change pattern and the timing of the event, the information providing unit that can extract the event that affects the trend change from the event information and provide the information. Have.
　本発明によれば、ユーザの健康状態の変化に関連するイベントを提示できるようにした情報伝達装置および情報伝達方法を提供することができる。 According to the present invention, it is possible to provide an information transmission device and an information transmission method capable of presenting an event related to a change in a user's health condition.
本発明の一実施形態に係る情報伝達システムの構成を示すブロック図である。It is a block diagram which shows the structure of the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおけるデータベースのデータ構造を示す図である。It is a figure which shows the data structure of the database in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、対象者の検査データの時系列的変化を示すグラフである。It is a graph which shows the time-series change of the inspection data of a subject in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、検査結果送信の動作を示すフローチャートである。It is a flowchart which shows the operation of the inspection result transmission in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、推奨施設の推論の動作を示すフローチャートである。It is a flowchart which shows the operation of the inference of the recommended facility in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、関連施設情報の取得の動作を示すフローチャートである。It is a flowchart which shows the operation of acquisition of the related facility information in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、検査結果送信の動作の変形例を示すフローチャートである。It is a flowchart which shows the modification of the operation of the inspection result transmission in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、影響イベントによって対象者の検査データが変化する様子を示すグラフである。It is a graph which shows how the inspection data of a subject changes by an influence event in the information transmission system which concerns on one Embodiment of this invention. 本発明の一実施形態に係る情報伝達システムにおいて、推論モデル作成の動作の変形例を示すフローチャートである。It is a flowchart which shows the modification of the operation of making an inference model in the information transmission system which concerns on one Embodiment of this invention.
　以下、本発明の一実施形態として、情報伝達システムに本発明を適用した例について説明する。本実施形態においては、対象者の状況を考慮することによって正確な健康状態を把握し、カスタマイズ情報を提供する例として、対象者のプロフィール情報を記憶しておき、日々、健康状態に関する検査データをモニタリングし、これらの情報に基づいて、正確な健康状態を把握するための検査ができる施設、および／または治療ができる施設の情報を提供することが可能な情報伝達装置および情報伝達方法を説明する。この実施形態における対象者は、再検査によっては、患者となるかもしれない者である。また、この対象者は、再検査の結果によっては、健康に自信を取り戻し、病気の心配をせず、さらに日々の暮らしを楽しむことができる者でもある。また、簡単な生活の改善や治療などで、同様に健康体となることが出来る者でもある。 Hereinafter, an example in which the present invention is applied to an information transmission system will be described as an embodiment of the present invention. In the present embodiment, as an example of grasping the accurate health condition by considering the situation of the subject and providing customized information, the profile information of the subject is stored, and the inspection data regarding the health condition is collected every day. Explain information transmission devices and information transmission methods that can provide information on facilities that can be monitored and / or can be treated based on this information. .. The subject in this embodiment is a person who may become a patient, depending on the re-examination. In addition, depending on the results of the re-examination, this subject is also a person who can regain confidence in health, do not have to worry about illness, and enjoy daily life. In addition, he is also a person who can become healthy by simple improvement and treatment of life.
　本実施形態に係る情報伝達システムは、対象者の検査データを取得する検査データ取得部（例えば、図１の情報判定機器２参照）と、対象者のプロフィール情報と医療機関ごとの保有機器情報を記憶する記憶部（例えば、図１のＤＢ部８参照）と、検査データ、対象者のプロフィール情報、および医療機関ごとの保有機器情報に従って、対象者への伝達情報を決定する伝達情報決定部（例えば、図１の制御部１参照）を有する。この情報伝達システムは、例えば、サーバによって構成されるが、サーバと情報のやり取りが可能なパーソナルコンピュータ、スマートフォン等の携帯情報機器等によって構成してもよい。 The information transmission system according to the present embodiment has an inspection data acquisition unit (for example, refer to the information determination device 2 in FIG. 1) that acquires the inspection data of the subject, profile information of the subject, and information on the possessed equipment for each medical institution. A storage unit (for example, see DB 8 in FIG. 1) to be stored, and a transmission information determination unit (for example, a transmission information determination unit) that determines transmission information to the target person according to test data, profile information of the target person, and possessed device information for each medical institution. For example, it has a control unit 1) in FIG. This information transmission system is composed of, for example, a server, but may be configured by a personal computer capable of exchanging information with the server, a mobile information device such as a smartphone, or the like.
　また、情報伝達システムは、この対象者の検査データを取得する検査データ取得部（例えば、図１の情報判定機器２参照）と、検査データの時系列的な変化パターンと対象者の医療機関への来院情報を用いて機械学習によって生成された推論モデルに従って、対象者への伝達情報を決定する伝達情報決定部（例えば、図１の制御部１、推論エンジン７参照）を有する。来院情報は、医療機関に限らず、検査機関において検査を受けた場合や、薬局等において医薬を購入し服薬した場合の情報も含まれる。この情報伝達システムも、例えば、サーバによって構成されるが、サーバと情報のやり取りが可能なパーソナルコンピュータ、スマートフォン等の携帯情報機器等によって構成してもよい。 In addition, the information transmission system is sent to the inspection data acquisition unit (for example, refer to the information determination device 2 in FIG. 1) that acquires the inspection data of the subject, the time-series change pattern of the inspection data, and the medical institution of the subject. It has a transmission information determination unit (see, for example, control unit 1 and inference engine 7 in FIG. 1) that determines transmission information to a subject according to an inference model generated by machine learning using the visit information of the above. The visit information is not limited to medical institutions, but also includes information when a person is examined at a medical institution or when a drug is purchased and taken at a pharmacy or the like. This information transmission system is also composed of, for example, a server, but may also be composed of a personal computer capable of exchanging information with the server, a mobile information device such as a smartphone, and the like.
　対象者への伝達情報を決定するにあたって、対象者の検査結果、この検査結果に基づいて更に検査等のために必要となる設備、この設備のある施設に基づいて、検索および／または推論を行う。この検索・推論を行うために、設備を有する施設を記憶するデータベース（ＤＢ）を設けておくとよい。また、検索・推論した結果に基づいて伝達情報を提供するにあたって、施設名や電話やメール、地図等を含むアクセス方法、診察時間や空き時間、費用概算等の情報を含んでもよい。また、施設は、一つに限らず、複数であってもよい。 In determining the information to be transmitted to the subject, search and / or infer based on the inspection result of the subject, the equipment required for further inspection, etc. based on this inspection result, and the facility with this equipment. .. In order to perform this search / inference, it is advisable to provide a database (DB) for storing facilities having facilities. In addition, in providing transmission information based on the results of search / inference, information such as facility name, access method including telephone, mail, map, etc., consultation time, free time, cost estimation, etc. may be included. Further, the number of facilities is not limited to one, and may be multiple.
　通常、医療機関に納入された医療機器の保守は納入した医療機器メーカーが担っており、施設や患者単位で設置機器が必要となるため、機器管理のサービス提供型へのサービス・ビジネスが形成されつつある。このようなサービスと連携すれば、本実施形態において使用するデータベースは、このサービスにあるものを利用してもよい。このサービスは、顧客情報も含めて一元管理し、診療科や様々な患者層、医師、看護師など顧客データに基づいて、設置機器の構成管理や変更管理機能を提供するため、サービスを構成するアイテム（構成アイテム）の登録や、設置機器の構成情報を逐次刷新することによって、バージョンアップや部品交換時の影響範囲が管理しやすくなっている。このためリプレース時期の予測することや、設置機器の品質維持が可能になる。ここでは医療機関について述べたが、それ例外にも、医療機関における診療等も含む、健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報も、同様に、上述のサービスの中で一元管理されることを想定している。 Normally, the maintenance of medical equipment delivered to medical institutions is carried out by the medical equipment manufacturer that delivered it, and installation equipment is required for each facility or patient, so a service business for equipment management service provision type is formed. It's getting better. By linking with such a service, the database used in this embodiment may be the one in this service. This service is configured to centrally manage including customer information and provide configuration management and change management functions for installed equipment based on customer data such as clinical departments, various patient groups, doctors, and nurses. By registering items (configuration items) and sequentially updating the configuration information of installed equipment, it is easier to manage the range of influence when upgrading or replacing parts. Therefore, it is possible to predict the replacement time and maintain the quality of the installed equipment. Although medical institutions have been described here, as an exception, information on equipment and / or equipment and / or environment at the time of a health response event, including medical treatment at medical institutions, is also described above. It is assumed that it will be centrally managed within the service of.
　なお、健康対応イベントは、健康に影響を与えたようなイベントであり、例えば、ユーザが医療施設に行って医師による診察を受けたことや、薬局で薬を購入し服薬した等、健康に関係する行為である。また、ユーザがジムに行きトレーニングを行ったことや、テニス等の運動を行ったことや、更に、飲食店において暴飲暴食したことや、睡眠不足になったことや、寒中で仕事や遊んだこと等、健康に直接または間接に影響を与えるような行為であればよい。 A health response event is an event that has an impact on health. For example, a user goes to a medical facility to see a doctor, or purchases and takes medicine at a pharmacy, which is related to health. It is an act to do. In addition, users went to the gym for training, exercised tennis, etc., and also had binge eating at restaurants, lacked sleep, and worked or played in the cold. Any act that directly or indirectly affects health, such as.
　また、推論エンジンを用いて、推論によって対象者への伝達情報を決定するようにしてもよい。この場合には、対象者の検査データを推論エンジンに入力し、対象者への伝達情報を取得する。このとき、推論結果の信頼性を判定し、信頼性が低い場合には、教師データを収集して推論モデルを生成のための学習を依頼する（例えば、図７のＳ８、Ｓ１０等参照）。教師データは推論エンジンへの入力データと類似のデータを入手して作成する。新たな推論モデルが生成されれば、この推論モデルを用いて、対象者への伝達情報を決定する。 Alternatively, an inference engine may be used to determine the information to be transmitted to the target person by inference. In this case, the inspection data of the target person is input to the inference engine, and the information transmitted to the target person is acquired. At this time, the reliability of the inference result is determined, and if the reliability is low, teacher data is collected and learning for generating an inference model is requested (see, for example, S8, S10, etc. in FIG. 7). Teacher data is created by obtaining data similar to the data input to the inference engine. When a new inference model is generated, this inference model is used to determine the information to be transmitted to the target person.
　また、健康状態に影響を与えるようなイベント（影響イベント）が発生すると、時系列的な検査データに変化が生じる（例えば、図８（ａ）参照）。このような環境変化などを含む影響イベントは、他のユーザ等の時系列的な検査データを用いて、推論モデルを生成し、この推論モデルを用いれば、ユーザの検査データから推論することができる。すなわち、特定の仕様の検査機器を用いて特定期間の時系列的に主に生体情報を検査する検査データを取得して、その変化を時系列に並べると検査データの中に、刻々と変化する変化パターンが得られる。生体情報の数値が健康状態や体調の変化で変わる場合、その数値の変化をみれば、その人の健康状態や体調の変化の傾向が得られる。この数値が時間的な経過を経ても一定であれば、その数値から健康状態は一定の状態を保っていると考えられる。また、生活のリズムや季節や気候の影響や成長や加齢等によって緩やかな変化がある場合には、健康であってもありうるので問題にしなくて良い。すなわち、検査データの時間に従った変化パターンは緩やかであったり、規則的であったりしながらも、略一定である。 In addition, when an event that affects the health condition (effect event) occurs, the time-series test data changes (see, for example, FIG. 8 (a)). Impact events including such environmental changes can be inferred from the user's inspection data by generating an inference model using the time-series inspection data of other users and the like. .. That is, if test data for mainly inspecting biological information is acquired in a time series for a specific period using a test device having a specific specification and the changes are arranged in a time series, the test data changes every moment. A change pattern is obtained. When the numerical value of the biological information changes due to the change of the health condition or the physical condition, the tendency of the change of the health condition or the physical condition of the person can be obtained by observing the change of the numerical value. If this value is constant over time, it is considered that the health condition is kept constant from the value. In addition, if there is a gradual change due to the rhythm of life, the influence of the season or climate, growth, aging, etc., it may be healthy, so it does not matter. That is, the time-dependent change pattern of the inspection data is gradual or regular, but is substantially constant.
　なお、生体情報である以上、測定値は何らかの変化をする。ここでパターンと書いたものが連続する微小なデータ変化だとすると、パターンが変化するという表現は、測定値が微小変化しながらも傾向（トレンド）を持つ場合、そのトレンドが変化している場合と、前述の意味のパターンが変化している場合において、両者の表現を区別しないと、混乱をきたす。そこで、微小な変動をしながらも、それが上昇したり下降したり、周期がある場合、それが通常から変わる状況を表す言葉を、傾向（トレンド）変化と言い直してもよい。 As long as it is biometric information, the measured value will change in some way. If what is written as a pattern here is a continuous minute change in data, the expression that the pattern changes means that the measured value has a tendency (trend) even though it changes slightly, that the trend is changing, and that the trend is changing. When the above-mentioned pattern of meaning is changing, it is confusing if the two expressions are not distinguished. Therefore, if there is a cycle in which the fluctuation is small, but it rises or falls, and there is a cycle, the term that expresses the situation in which it changes from normal may be rephrased as a trend change.
　つまり、一定の健康状態であれば、パターン（特定の値を中心とした微小な変動）も一定で、そのトレンドも一定と言ってもよい。ただし、このような微小な変化での変化パターンではなく、数値がそれまでの傾向とは異なるような値になったり、変化傾きや変動幅などで変動して、それまでの傾向（トレンド）に戻らなかったりする場合（特異な変化で別の傾向に推移する場合）、その人に何らかの健康状態や体調の変化があったと考えられる。この時、パターンのトレンドが変わったとする。 In other words, if you are in a certain state of health, you can say that the pattern (small fluctuations centered on a specific value) is also constant, and the trend is also constant. However, instead of the change pattern due to such a small change, the numerical value may be different from the previous tendency, or it may fluctuate depending on the change slope or fluctuation range, and the tendency (trend) up to that point may be changed. If it does not return (if it changes to a different trend due to a peculiar change), it is probable that the person had some change in health or physical condition. At this time, it is assumed that the trend of the pattern has changed.
　このような変化パターン（これは定常的な変化）の傾向をモニタしておき、一日や一週間、一ヵ月単位などの特定期間での傾向（最大値や最小値や平均値や変動周期や変動幅の傾向）を記録、把握しておく。そして、新しく得られた検査結果を集めてパターンとし、このパターンの個々の検査数値を比較し、このパターンが持つ傾向が、これまでの傾向（トレンド）が変わったかを判定すればよい。もちろん、大きく傾向が違った数値は出れば、明らかにトレンドと違うと考えられる。また、一回の比較である必要はなく、その傾向変化が続く場合など、まとまったデータ群のパターンの傾向で判定してもよい。 The tendency of such a change pattern (this is a steady change) is monitored, and the tendency (maximum value, minimum value, average value, fluctuation cycle, etc.) in a specific period such as one day, one week, or one month unit. Record and understand the tendency of fluctuation range). Then, the newly obtained test results may be collected into a pattern, and the individual test values of this pattern may be compared to determine whether the tendency of this pattern has changed the tendency (trend) so far. Of course, if a numerical value with a significantly different trend appears, it is clearly considered to be different from the trend. Further, it is not necessary to make a single comparison, and it may be determined by the tendency of the pattern of a cohesive data group, such as when the tendency change continues.
　パターンの傾向変化がある場合には、パターンの傾向に変化を引き起こした、きっかけとなるイベントがあると考えられる。例えば、体温が３６℃近傍で変化するだけの人の体温が、急に３８℃になる場合などは、その時点で変化ありとしてもよい。また、この場合には、体力が低下して、保菌者と接触した、というような、それまでと違った出来事があったとも考えられる。 If there is a change in the pattern tendency, it is considered that there is an event that triggered the change in the pattern tendency. For example, when the body temperature of a person whose body temperature only changes in the vicinity of 36 ° C suddenly becomes 38 ° C, there may be a change at that time. In this case, it is also probable that there was a different event, such as a decrease in physical strength and contact with a carrier.
　このように、変化パターンの傾向変化があった際のイベントは、体調や健康に影響しているので、仮に影響イベントと呼ぶ。特定期間内で変化パターンの傾向が変化した時点（大義的にはパターンが変化したと表現してもよいので、煩雑さを避けるために以下の実施例中には、パターンのトレンドが変化したことをパターン変化と現すこともある）に対応する時点における影響イベント情報を検出すること（記録しておいたものを探し出したり、その時点における外部情報を検索したりしてもよい）は、その人に限らず、他の人にとっても健康管理上、重要な情報となり得る。検出された影響イベント情報をアノテーションした結果を、健康管理用のイベント推論用の推論モデル学習に用いる教師データとして得ることが出来る。 In this way, the event when there is a change in the tendency of the change pattern affects the physical condition and health, so it is tentatively called an influence event. When the trend of the change pattern changes within a specific period (in a broad sense, it may be expressed that the pattern has changed, so in order to avoid complication, the trend of the pattern has changed during the following examples. Detecting the impact event information at the time corresponding to (sometimes expressed as a pattern change) (you may search for the recorded one or search for external information at that time) is the person. It can be important information for health management not only for other people but also for other people. The result of annotating the detected influence event information can be obtained as teacher data used for inference model learning for event inference for health management.
　このような教師データで学習させた推論モデルを取得し、特定の人物の特定の仕様の検査機器を用いて特定期間の幅と類似の期間で時系列的に得た検査データを、推論モデルに入力して影響イベントを推論し、該推論の結果を伝達することが出来る。ここで特定期間の幅と書いたのは、何年もかけて得たデータの変化と、一週間で得たデータ変化は、同じ人間同士の検査データの比較である場合、正しい比較とならないからで、朝昼晩、あるいは起床時、運動時などで変わる数値の場合にそれらの影響を考慮した変化パターン（特定の値を中心とした微小な変動）の特徴までを、正しく捉えたパターンとして比較するためである。つまり、検査機器を用いて取得した特定期間の時系列的な変化パターンを他の人の傾向と比較等した場合、時系列的な検査データが変化していた場合に、この変化が生じたことに影響イベント情報を検出し、また推論することによって有益な情報を取得するシステムにおいては、特定期間を類似の期間として比較した方が良い。 An inference model trained from such teacher data is acquired, and the inspection data obtained in time series in a period similar to the width of a specific period using an inspection device of a specific specification of a specific person is used as an inference model. You can input to infer the influence event and convey the result of the inference. The reason why I wrote the width of a specific period here is that the change in data obtained over many years and the change in data obtained in one week are not correct comparisons when comparing test data between the same people. So, in the case of numerical values that change in the morning, day and night, when waking up, when exercising, etc., even the characteristics of the change pattern (small fluctuation centered on a specific value) that considers those effects are compared as a pattern that correctly captures To do. In other words, when comparing the time-series change pattern of a specific period acquired using inspection equipment with the tendency of other people, this change occurred when the time-series inspection data had changed. In a system that obtains useful information by detecting and inferring event information, it is better to compare specific periods as similar periods.
　上述の期間は、以下のような考え方で選択するとよい。
（１）慢性的に生体情報が変化する疾病特有の生体情報など、緊急対応を必要としない可能性が高い場合：情報取得時の上述の類似期間は、データベースや推論モデルの推奨された仕様に従って決まる特定期間に相当する期間であって、今後の検査結果が集まるまで待つ。
（２）急性の疾病特有の生体情報で、緊急対応を必要とするかもしれない場合：情報取得時の上述の類似期間は、既にその情報を受ける人、あるいは介護者や関係者などが取得した期間に相当する期間に類似する期間の時系列データ（かつ類似の傾向を持つものを選んでもよい）をデータベースから選んで使う。また、そのような期間のデータを使った推論モデルを選ぶか、そのデータベースにある類似期間のデータを教師データ化によって新たに推論モデルを作成、それを使う。 The above period may be selected based on the following concept.
(1) When there is a high possibility that emergency response is not required, such as disease-specific biometric information whose biometric information changes chronically: The above-mentioned similar period at the time of information acquisition follows the recommended specifications of the database and inference model. It is a period corresponding to a specific period to be decided, and waits until future test results are collected.
(2) When the biometric information peculiar to an acute disease may require emergency response: The above-mentioned similar period at the time of information acquisition has already been acquired by the person who has already received the information, or a caregiver or a person concerned. Time-series data of a period similar to the period corresponding to the period (and may be selected with a similar tendency) is selected from the database and used. Also, select an inference model that uses the data of such a period, or create a new inference model by converting the data of a similar period in the database into teacher data and use it.
　ただし、これらの期間は厳密である必要はない。前者（上述の（１））の場合、１年単位で測定することが推奨されていても、１０か月分のデータしかない場合に、不足している２か月分のデータは傾向から予測して追加して利用してもよい。一方、後者（上述の（２））の場合、今日と昨日の２日分のデータしかなくとも、例えば、昨日のデータが複数回のデータがあって安定しているような場合は、その前日も同様の変化だったとして補足して利用してもよい。その情報を受ける人、あるいは介護者や関係者などの要望で、今、知りたいというタイミング（例えば起床時とか出かける前であって、これからの行動を決めなければならない場合など）があるので、その場合は、上述したような補完情報を利用してもよい。ただし、この場合には、データベースなり推論モデルなりが有する、データの種別別や疾病別の仕様に従って、それにほぼ合わせた期間にする。 However, these periods do not have to be strict. In the case of the former ((1) above), even if it is recommended to measure in units of one year, if there is only 10 months'worth of data, the missing 2 months' worth of data is predicted from the tendency. You may add and use it. On the other hand, in the latter case ((2) above), even if there are only two days' worth of data today and yesterday, for example, if yesterday's data is stable with multiple data, the day before that. May be supplemented and used as the same change. At the request of the person who receives the information, the caregiver, the person concerned, etc., there is a timing that I want to know now (for example, when I wake up or before going out and I have to decide what to do in the future), so that In that case, the complementary information as described above may be used. However, in this case, according to the specifications for each type of data and each disease that the database or inference model has, the period is set to be approximately the same.
　また、ここで想定するデータベースは、イベントによって、検査データの変化パターンのトレンドが変化する、というイベント情報を含むことを想定しており、健康に影響するイベントごとに、データベースが分けられていてもよい。例えば、慢性疾患の数値が、特定の薬の服用によって改善したり、特定の生活習慣で悪化したり、といった様子が表されたデータベースであることが好ましい。感染症に罹患した時の罹患時の状況と潜伏期間を経ての発症までの変化などが分かるようにしてもよく（これは潜伏期間が長いものは判定困難であるが、推定原因などが情報として盛り込まれていることが好ましい）、この発症時の発熱によって体温の日々の変動の傾向が乱れ、急上昇する等、傾向が変わった時のトレンド変化の様子を記録したものであることが好ましい。これによって、検査結果のトレンド変化とその原因が推定でき、情報を受け取った人は、その情報で、健康維持や改善や悪化の防止に留意が可能となる。　 In addition, the database assumed here is assumed to include event information that the trend of the change pattern of the test data changes depending on the event, and even if the database is divided for each event that affects health. Good. For example, it is preferable to use a database that shows how the numerical values of chronic diseases are improved by taking a specific drug or worsened by a specific lifestyle. It may be possible to understand the situation at the time of illness when infectious disease and the change from the onset after the incubation period (this is difficult to determine if the incubation period is long, but the probable cause etc. can be used as information. It is preferable that it is included), and it is preferable to record the state of the trend change when the tendency changes, such as the tendency of daily fluctuation of body temperature being disturbed and suddenly rising due to the fever at the time of onset. As a result, the trend change of the test result and its cause can be estimated, and the person who receives the information can pay attention to the maintenance of health and the prevention of improvement and deterioration.
　なお、推論モデルを生成するためには、適正規模の教師データからなる学習用母集団を作成しなければならない。このために、多数の個人の情報を収集して学習用母集団を作成する。この場合、検査機器の種類や、また検査を行った期間等の情報を含めて収集し、これらの情報毎に学習用母集団を作成するとよい。 In addition, in order to generate an inference model, it is necessary to create a learning population consisting of teacher data of an appropriate scale. For this purpose, a large number of individual information is collected to create a learning population. In this case, it is advisable to collect information such as the type of inspection equipment and the period during which the inspection was performed, and create a learning population for each of these pieces of information.
　次に、図１を用いて、本発明の一実施形態にかかわる情報伝達システムの構成を説明する。この情報伝達システムは、制御部１、情報判定機器２、端末４、学習部５、推論エンジン７、データベース（ＤＢ）部８、関連検査機関（医療機関等を含む）９とからなる。なお、このデータベースは連携した複数のデータベースやコンピュータが介在するものであってもよい。これらの各部の内、制御部１は、サーバ内に配置され、情報判定機器２、端末４、学習部５、推論エンジン７、ＤＢ部８、関連検査機関９は、インターネット等のネットワークを通じてサーバに接続可能としている。しかし、本実施形態は、この構成に限定されることなく、例えば、制御部１、情報判定機器２、学習部５、推論エンジン７、ＤＢ部８の内のいずれか１つまたは複数が、サーバ内に配置され、他は別のサーバやパーソナルコンピュータ等の電子機器に、配置されていてもよい。更に、関連検査機関９が、サーバの機能を有してもよい。 Next, the configuration of the information transmission system according to the embodiment of the present invention will be described with reference to FIG. This information transmission system includes a control unit 1, an information determination device 2, a terminal 4, a learning unit 5, an inference engine 7, a database (DB) unit 8, and related inspection institutions (including medical institutions) 9. It should be noted that this database may be mediated by a plurality of linked databases and computers. Of these units, the control unit 1 is arranged in the server, and the information determination device 2, the terminal 4, the learning unit 5, the inference engine 7, the DB unit 8, and the related inspection organization 9 are connected to the server via a network such as the Internet. It is possible to connect. However, the present embodiment is not limited to this configuration, and for example, any one or more of the control unit 1, the information determination device 2, the learning unit 5, the inference engine 7, and the DB unit 8 may be a server. It may be arranged inside, and the others may be arranged in another electronic device such as a server or a personal computer. Further, the related inspection agency 9 may have a server function.
　制御部１は、本実施形態に係る情報伝達システムを制御するコントローラであり、サーバ等や、ネットワークを通じて他の端末にファイルやデータなどを提供するＣＰＵ（Central Processor Unit）、メモリ、ＨＤＤ（Hard Disc Drive）等から構成されているＩＴ機器を想定している。しかし、制御部１は、この構成に限らず、小規模なシステムとして構築する場合は、パーソナルコンピュータのようなものでも構成は可能である。制御部１は、各種のインターフェース回路を有し、他の機器と連携することができ、プログラムによってさまざまな演算制御が可能である。 The control unit 1 is a controller that controls an information transmission system according to the present embodiment, and is a CPU (Central Processor Unit), a memory, and an HDD (Hard Disc) that provide files, data, and the like to a server and the like and other terminals via a network. It is assumed that the IT equipment is composed of Drive) and the like. However, the control unit 1 is not limited to this configuration, and when it is constructed as a small-scale system, it can be configured with something like a personal computer. The control unit 1 has various interface circuits, can cooperate with other devices, and can perform various arithmetic controls by a program.
　制御部１は、連携する各装置から情報を受け取り、情報を整理し、必要な情報を生み出し、この情報をユーザに提供する。制御部１は、連携する各装置に依頼を出力し、また各装置を操作するような機能も有している。本実施形態においては、システムの自由度の高さや使い勝手を想定し、情報判定機器２や対象者が有する端末４等と制御部１の間は、無線通信や有線通信で接続可能となっている。このための通信としては、無線ＬＡＮや携帯電話通信網を想定し、状況に応じてブルートゥース（登録商標）や赤外通信などの近距離無線などを併用してもよい。通信回路、アンテナや接続端子等からなる通信部の記載は煩雑になるので図１においては省略してあるが、図中の通信を示す矢印の部分には、通信回路等を有する通信部が設けられている。 The control unit 1 receives information from each linked device, organizes the information, generates necessary information, and provides this information to the user. The control unit 1 also has a function of outputting a request to each of the linked devices and operating each device. In the present embodiment, assuming a high degree of freedom and usability of the system, it is possible to connect the information determination device 2, the terminal 4 and the like owned by the target person, and the control unit 1 by wireless communication or wired communication. .. As the communication for this purpose, a wireless LAN or a mobile phone communication network is assumed, and short-range wireless communication such as Bluetooth (registered trademark) or infrared communication may be used in combination depending on the situation. The description of the communication unit including the communication circuit, the antenna, the connection terminal, etc. is complicated, so it is omitted in FIG. 1, but the communication unit having the communication circuit, etc. is provided in the part of the arrow indicating the communication in the figure. Has been done.
　制御部１は、通信制御部１ａ、ＩＤ判定部１ｂ、情報提供部１ｃ、推論モデル仕様決定部１ｄ、推論依頼部１ｅ、検索部１ｆを有する。これらの各部は、制御部１内のＣＰＵおよびプログラム等による、ソフトウエアによって実現してもよく、またハードウエア回路によって実現してもよく、またソフトウエアとハードウエア回路を協働させることによって実現してもよい。また、図１において、制御部１内の各部は、互いに連携してそれぞれの機能を果たすため信号の方向は省略しているが、これは、別途、フローチャートで説明する。例えば、図４のＳ１のようなステップにおいて、ＩＤ判定部１ｂは、情報判定機器２等から、同一の対象者毎に情報を収集している。 The control unit 1 includes a communication control unit 1a, an ID determination unit 1b, an information provision unit 1c, an inference model specification determination unit 1d, an inference request unit 1e, and a search unit 1f. Each of these parts may be realized by software by a CPU, a program, or the like in the control unit 1, may be realized by a hardware circuit, or may be realized by coordinating software and a hardware circuit. You may. Further, in FIG. 1, since each unit in the control unit 1 cooperates with each other to perform each function, the direction of the signal is omitted, but this will be described separately with a flowchart. For example, in a step like S1 in FIG. 4, the ID determination unit 1b collects information from the information determination device 2 and the like for each of the same target persons.
　通信制御部１ａは、通信回路等を有し、情報判定機器２、端末４、学習部５、推論エンジン７、データベース（ＤＢ）部８、および関連検査機関９内に設けられた通信部と、データ等の送受信を行う。情報判定機器２、端末４等の各機器・各部もそれぞれ通信部を有しているが、図１においては煩雑になるため、図示を省略している。 The communication control unit 1a has a communication circuit and the like, and includes an information determination device 2, a terminal 4, a learning unit 5, an inference engine 7, a database (DB) unit 8, and a communication unit provided in the related inspection organization 9. Send and receive data, etc. Each device / part such as the information determination device 2 and the terminal 4 also has a communication unit, but the illustration is omitted in FIG. 1 because it is complicated.
　ＩＤ判定部１ｂは、情報判定機器２等から、同一の対象者毎に情報を収集する。情報判定機器２によって情報が取得された個人を特定するため、個人毎にＩＤが割り当てられている。本実施形態においては、ユーザ個々のデータを取り扱うので、どのユーザの情報を受け取って、どのユーザにガイドを出すかの管理は、ＩＤ判定部１ｂが行っている。この特定ユーザの判定は、情報判定機器２が生体認証機能を有したり、ユーザが端末４によって情報判定機器２内の通信部を通じて通信したり、また端末４が固有のコードを読み取ったりすることによって行う。なお、個人情報を保護するために、必要な部分を暗号化し管理を厳しくするが、これらは汎用的な技術であることから、詳しい説明を省略する。 The ID determination unit 1b collects information from the information determination device 2 and the like for each of the same target persons. An ID is assigned to each individual in order to identify the individual whose information has been acquired by the information determination device 2. In the present embodiment, since the data of each user is handled, the ID determination unit 1b manages which user's information is received and which user is given the guide. The determination of the specific user is that the information determination device 2 has a biometric authentication function, the user communicates with the terminal 4 through the communication unit in the information determination device 2, and the terminal 4 reads a unique code. Do by. In addition, in order to protect personal information, necessary parts are encrypted and management is strict, but since these are general-purpose technologies, detailed description will be omitted.
　情報提供部１ｃは、ユーザに正しい情報を提供するために、ユーザの情報を取得（他の装置が取得してあった結果を参照してもよい）する機能を有する。また、情報提供部１ｃは、情報判定機器２等から取得したユーザ（ＩＤによって特定される）の検査データや、関連検査機関９から取得した種々の情報、およびＤＢ部８に記憶された保有機器に関する情報やユーザのプロフィール情報等を用いて、ユーザの健康状態を判断し、必要に応じて検査や治療を受けるべき施設に関する情報が決まると、この情報をユーザに提供する。 The information providing unit 1c has a function of acquiring user information (may refer to the result acquired by another device) in order to provide correct information to the user. In addition, the information providing unit 1c includes inspection data of the user (specified by the ID) acquired from the information determination device 2 and the like, various information acquired from the related inspection organization 9, and a possessed device stored in the DB unit 8. When the information about the user and the profile information of the user are used to judge the health condition of the user and the information about the facility to be examined or treated is determined as necessary, this information is provided to the user.
　すなわち、情報提供部１ｃは、ユーザに検査や治療を受けるに適した施設を推奨するための情報を提供する。情報提供部１ｃは、情報判定機器２から送信されてきた検査データを入力する。このデータは、後述するように、時間情報が付された検査データ（時系列情報）であり、図３に示すようなグラフにできるようなデータ構造でＤＢ部８等に蓄積される。なお、本実施形態においては、情報判定機器２からの情報を用いて制御部１がユーザへ情報提供を行うことを想定しているが、関連検査機関９を有するサーバが、同様に情報を収集するような変形例であってもよい。 That is, the information providing unit 1c provides the user with information for recommending a facility suitable for receiving an examination or treatment. The information providing unit 1c inputs the inspection data transmitted from the information determination device 2. As will be described later, this data is inspection data (time series information) to which time information is attached, and is stored in the DB unit 8 or the like with a data structure that can be graphed as shown in FIG. In the present embodiment, it is assumed that the control unit 1 provides information to the user using the information from the information determination device 2, but the server having the related inspection organization 9 collects the information in the same manner. It may be a modified example such that
　また、情報提供部１ｃは、ユーザの住所や勤務する場所における行動様式や食生活や就寝時間や食事のタイミングなど生活習慣等を、インターネット上において取得してもよく、この取得した情報も加味して、ユーザに提供する施設等の情報を生成してもよい。これらの情報の取得は、汎用または広く知られた技術で補完が可能である。これらの情報を取得することによって生成した施設等の情報のカスタマイズも、また、情報提供部１ｃが行ってもよい。具体的には、ユーザの住居（ユーザのプロフィール情報として記録されている）の近くにあるクリニックの情報などを提供することが考えられる。しかし、そのクリニックに肝心の検査の設備がない場合には、疾病の原因推定と対策、治療などが出来ない。そこで、クリニックの有する設備や勤務する医師の専門、その他の施設に関連するプロフィール情報を加味して、情報提供を行う。この施設に関するプロフィール情報は、関連検査機関９から医療機関情報として取得する。 In addition, the information providing unit 1c may acquire lifestyle habits such as the user's address, behavioral style at the place of work, eating habits, bedtime, and meal timing on the Internet, and the acquired information is also taken into consideration. Therefore, information such as facilities to be provided to the user may be generated. The acquisition of this information can be complemented by general-purpose or well-known technology. The information providing unit 1c may also customize the information of the facility or the like generated by acquiring the information. Specifically, it is conceivable to provide information on a clinic near the user's residence (recorded as the user's profile information). However, if the clinic does not have the essential inspection equipment, it is not possible to estimate the cause of the disease, take countermeasures, or treat it. Therefore, we will provide information by adding profile information related to the facilities of the clinic, the specialty of the doctors who work, and other facilities. Profile information about this facility is obtained as medical institution information from the related inspection institution 9.
　情報提供部１ｃは、推奨施設等の情報提供に当たって、情報判定機器２、関連検査機関９から収集した情報に加えて、ＤＢ部８に記憶されている保有機器等の情報も利用する。もちろん、このＤＢ８の中に記録された情報は、ＤＢ部８以外の異なる記録部に記録されたものであってもよい。この場合には、図１におけるＤＢ部は複数となるが、煩雑になるので省略している。情報提供部１ｃは、情報を提供するにあたって、種々の情報を集める。すなわち、情報提供部１ｃは、対象者の検査データと、対象者のプロフィール情報と、検査・医療機関ごとの保有機器情報を取得する取得部として機能する。 In providing information on recommended facilities, etc., the information providing unit 1c uses the information on the possessed equipment, etc. stored in the DB unit 8 in addition to the information collected from the information judgment device 2 and the related inspection organization 9. Of course, the information recorded in the DB 8 may be recorded in a different recording unit other than the DB unit 8. In this case, there are a plurality of DB parts in FIG. 1, but they are omitted because they are complicated. The information providing unit 1c collects various kinds of information in providing the information. That is, the information providing unit 1c functions as an acquisition unit for acquiring the test data of the target person, the profile information of the target person, and the possessed device information for each test / medical institution.
　情報提供部１ｃは、情報判定機器２、関連検査機関９から送信されてきたユーザの検査データの中に、特定情報があるか否かを判定し、特定情報を検出した場合には、さらに検査を行うための推奨施設を提示する（図４のＳ１、Ｓ３、Ｓ１３、図７のＳ１、Ｓ３、Ｓ１３ａ参照）。特定情報は、疾病と関連する情報であり、例えば、健康な状態と差異がある数値や、変化パターン（特定の値を中心とした微小な変動）のトレンド（の変化）である。特定情報の値そのものが標準値から大きく外れている場合や、変化パターンのトレンド変化が顕著な場合には、何らかの疾病がある可能性が高い。特定情報は、特定の疾患を疑うことが出来る程度の情報であればよい。 The information providing unit 1c determines whether or not there is specific information in the user's inspection data transmitted from the information determination device 2 and the related inspection organization 9, and if the specific information is detected, further inspection is performed. (See S1, S3, S13 in FIG. 4, S1, S3, S13a in FIG. 7). The specific information is information related to a disease, for example, a numerical value having a difference from a healthy state or a trend (change) of a change pattern (a minute change centered on a specific value). If the value of the specific information itself deviates significantly from the standard value, or if the trend change of the change pattern is remarkable, there is a high possibility that there is some kind of disease. The specific information may be information that can be suspected of a specific disease.
　特定情報を検出した場合に、推奨施設を提示するにあたって、ＤＢ部８に記録されている施設を検索するようにしてもよい（図４参照）。このために、ＤＢ部８には、特定情報毎や疾患毎に、必要となる検査や、これらの検査を行うための検査機器・設備等を有する医療施設や検査機関等のデータベースを構築しておけばよい。また、ＤＢ部８の検索する方法以外にも、推論エンジンによってユーザへのアドバイスを推論するようにしてもよい（図７参照）。この場合には、ユーザの履歴データ等を入力した場合に、ユーザの疾患や、必要となる検査や、推奨施設等を推論する推論モデルを設定した推論エンジンを利用すればよい。 When specific information is detected, the facility recorded in the DB section 8 may be searched when presenting the recommended facility (see FIG. 4). For this purpose, the DB unit 8 is constructed with a database of necessary tests for each specific information or disease, and a database of medical facilities, testing institutions, etc. having testing equipment / equipment for performing these tests. Just leave it. In addition to the search method of the DB unit 8, an inference engine may be used to infer advice to the user (see FIG. 7). In this case, when the user's history data or the like is input, an inference engine in which an inference model for inferring the user's disease, necessary tests, recommended facilities, etc. may be used may be used.
　情報提供部１ｃは、対象者の検査データと、プロフィール情報と、検査・医療機関ごとの保有機器情報に従って、対象者に伝達する伝達情報を決定する伝達情報決定部として機能する（図４のＳ７、Ｓ９、Ｓ１１、Ｓ１３等参照）。伝達情報決定部は、対象者の特定の時間幅で抽出された対象者の変化パターン（特定の値を中心とした微小な変動がその値をずらしながら変化するトレンド）情報を推論モデルに入力することにより推論結果を得て、この得られた推論結果に基づいて、伝達情報を決定する（例えば、図４のＳ９参照）。伝達情報決定部は、検査データに基づいて、必要となる検査を受けるための推奨医療機関に関する情報を、または対象者が検査・医療機関で受診するタイミングに関する情報を、伝達情報として決定する（図４のＳ７、Ｓ９、Ｓ１１、Ｓ１３、図３（ａ）のタイミングＴｃ等参照）。 The information providing unit 1c functions as a transmission information determination unit that determines the transmission information to be transmitted to the target person according to the test data of the target person, the profile information, and the possessed device information for each test / medical institution (S7 in FIG. 4). , S9, S11, S13, etc.). The transmission information determination unit inputs the change pattern (trend in which minute fluctuations centered on a specific value change while shifting the value) information of the target person extracted in a specific time width of the target person into the inference model. As a result, an inference result is obtained, and the transmission information is determined based on the obtained inference result (see, for example, S9 in FIG. 4). Based on the test data, the transmission information determination unit determines information on the recommended medical institution for undergoing the necessary examination, or information on the timing when the subject visits the examination / medical institution as transmission information (Fig.). 4 S7, S9, S11, S13, timing Tc and the like in FIG. 3A).
　上述の伝達情報決定部は、検査データ取得部（例えば、情報判定機器２）によって取得された検査データの変化パターンを推論部（例えば、推論エンジン７）に入力し、この推論部の推論結果に基づいて伝達情報を決定する（図５のＳ３５参照）。伝達情報決定部は、対象者に伝達する時系列パターンの抽出時から後の時点における伝達情報を決定する伝達情報とする。伝達情報決定部は、検査データの変化パターンを特定の時間幅で抽出し、この抽出した変化パターンを推論部に入力し、推論部から推論結果を取得する（例えば、図３（ａ）（ｂ）、図５のＳ３５参照）。 The transmission information determination unit described above inputs the change pattern of the inspection data acquired by the inspection data acquisition unit (for example, the information determination device 2) into the inference unit (for example, the inference engine 7), and uses the inference result of the inference unit as the inference result. The transmitted information is determined based on this (see S35 in FIG. 5). The transmission information determination unit is the transmission information that determines the transmission information at a time point after the extraction of the time-series pattern to be transmitted to the target person. The transmission information determination unit extracts the change pattern of the inspection data in a specific time width, inputs the extracted change pattern to the inference unit, and acquires the inference result from the inference unit (for example, FIGS. 3A and 3B). ), See S35 in FIG. 5).
　上述の伝達情報決定部は、検査データ取得部によって取得された検査データの変化パターンが特定の範囲内に収まっている場合に、推論部による推論を行う（例えば、図５のＳ２７、Ｓ３５参照）。伝達情報決定部は、検査データ取得部によって取得された検査データの変化パターンが特定の範囲外にある場合に、推論部による推論を行わない（例えば、図５のＳ２７、Ｓ２９参照）。伝達情報決定部は、対象者の検査データの変化パターンを推論部に入力することによって推論を行い、この推論結果に基づいて、対象者に伝達する伝達情報を決定する（例えば、図５のＳ３５参照）。情報伝達決定部は、対象者の検査データの時系列変化パターンを推論部に入力し、推論部が推論を行い、この推論結果に基づいて、対象者に伝達する時系列パターンから後の時時点における伝達情報を決定する。 The transmission information determination unit described above makes an inference by the inference unit when the change pattern of the inspection data acquired by the inspection data acquisition unit is within a specific range (see, for example, S27 and S35 in FIG. 5). .. The transmission information determination unit does not perform inference by the inference unit when the change pattern of the inspection data acquired by the inspection data acquisition unit is outside the specific range (see, for example, S27 and S29 in FIG. 5). The transmission information determination unit makes an inference by inputting a change pattern of the inspection data of the target person into the inference unit, and determines the transmission information to be transmitted to the target person based on the inference result (for example, S35 in FIG. 5). reference). The information transmission determination unit inputs the time-series change pattern of the inspection data of the subject into the inference unit, the inference unit makes an inference, and based on this inference result, the time-series pattern to be transmitted to the subject is later than the time point. Determine the information to be transmitted in.
　前述したように、検査データ取得部は、対象者の特定期間の時系列パターンとなる検査データを取得する。この取得する時系列パターンは、単に１回だけの測定によって得たデータではなく、複数の異なるタイミングに測定によって取得した個々の検査データによって構成され、検査データのパターンの変化までを情報として利用する。複数の検査データからなる時系列パターンを使用することよって、測定環境や状況の変化によって生ずる誤差の影響を受け難くしている。さらに、特定期間の終了時期から将来の時期（特定期間の延長時）における健康状態を推論し、将来に対する予測を可能にしている。 As described above, the inspection data acquisition unit acquires inspection data that is a time-series pattern of the target person for a specific period. This acquired time-series pattern is composed of individual inspection data acquired by measurement at a plurality of different timings, not simply data obtained by one-time measurement, and even changes in the inspection data pattern are used as information. .. By using a time-series pattern consisting of multiple inspection data, it is less susceptible to errors caused by changes in the measurement environment and conditions. Furthermore, it infers the health condition from the end of the specific period to the future period (when the specific period is extended), and makes it possible to predict the future.
　また、取得した時系列パターンに対して、対象者の検査・医療機関への来院のタイミング情報をアノテーション情報として付与すれば教師データができる。この教師データを用いて学習することによって生成された推論モデルを有する推論部があれば、特定期間（時系列変化パターンを取得するための期間）から先のタイミング（特定期間の延長時）に何が起こるかの推論ができる。なお、ここで使用される推論モデルを生成する際には、特定の入出力情報の仕様を規定し、学習を行う。 In addition, teacher data can be created by adding the timing information of the subject's examination / visit to the medical institution as annotation information to the acquired time-series pattern. If there is an inference part that has an inference model generated by learning using this teacher data, what is the timing (when the specific period is extended) beyond the specific period (period for acquiring the time series change pattern)? Can be inferred if When generating the inference model used here, the specifications of specific input / output information are specified and learning is performed.
　したがって、本実施形態においては、対象者の検査データの時系列変化パターンを、推論部に入力し、推論部が推論を行い、この推論結果に基づいて、特定期間から先のタイミングにおける伝達情報を決定する伝達情報決定部を設けている。このため、時系列パターンの検査取得時から先のタイミングにおける予測情報を伝達することができるシステム、装置、方法、プログラム等が提供できる。対象者の検査データは、個々の検査機器ごとに機械的な性能の差等があると、信頼性が低下していまう。そこで、同一タイプの検査機器（特定の仕様の検査機器）を用いて検査データの変化パターン情報を多数取得し、ビックデータとして扱えばよい。この場合、特定期間としては、固定の期間とする必要はなく、状況に応じて異なる時間幅（特定期間）としてもよい。なお、ここでいう「時間幅」は、測定タイミングと測定タイミングの間の時間（検査間隔・測定間隔）ではなく、一連の検査データの取得にあたって、最初の測定から最後の測定までの時間間隔をいう。「時間幅」は、この時間幅の中にたくさんの時系列データが含まれており、検査データの変化パターン情報を含む特定の時間幅と書き直してもよい。 Therefore, in the present embodiment, the time-series change pattern of the test data of the subject is input to the inference unit, the inference unit makes an inference, and based on this inference result, the transmitted information at the timing from the specific period to the future is transmitted. A transmission information determination unit for determining is provided. Therefore, it is possible to provide a system, an apparatus, a method, a program, or the like capable of transmitting the prediction information at the timing after the inspection acquisition of the time series pattern. The reliability of the inspection data of the subject will be reduced if there is a difference in mechanical performance for each inspection device. Therefore, a large amount of change pattern information of inspection data may be acquired by using the same type of inspection equipment (inspection equipment having specific specifications) and treated as big data. In this case, the specific period does not have to be a fixed period, and may be a different time width (specific period) depending on the situation. The "time width" here is not the time between measurement timings (inspection interval / measurement interval), but the time interval from the first measurement to the last measurement when acquiring a series of inspection data. Say. The "time width" may be rewritten as a specific time width including a lot of time series data in this time width and change pattern information of the inspection data.
　本実施形態において伝達情報決定部としての機能を果たす情報提供部１ｃは、学習部５によって生成された推論モデルが設定された推論エンジン７に、検査データの変化パターを入力し、アドバイスに関する推論結果を得て、入力された検査データに対応するユーザに提供する。このサービスは個人情報を利用する場合があり、アドバイス等の提供を受けるために個人情報の契約などが必要な場合がある。その意味で、ユーザのプロフィール情報が重要な場合もある。また、ユーザが幼児や高齢の場合は、そのユーザの世話をする人、介助者などにアドバイスを届けてもよい。これもユーザのプロフィール情報で管理した情報に従ってアドバイスなどの有効情報が届く。 The information providing unit 1c, which functions as a transmission information determining unit in the present embodiment, inputs a change putter of inspection data into the inference engine 7 in which the inference model generated by the learning unit 5 is set, and infers the result regarding advice. Is obtained and provided to the user corresponding to the input inspection data. This service may use personal information, and may require a contract for personal information in order to receive advice. In that sense, the user's profile information may be important. In addition, when the user is an infant or an elderly person, advice may be delivered to a person who takes care of the user, a caregiver, or the like. In this case as well, valid information such as advice arrives according to the information managed by the user's profile information.
　また、情報提供部１ｃは、特定の仕様の検査機器を用いて特定期間の時系列的に検査データを取得し、該検査データの変化パターン情報を取得するデータ取得部として機能する（例えば、図８（ｂ）のＳ６１、Ｓ６３参照）。情報提供部１ｃは、特定の健康対応イベントの入力に対応し、健康対応イベントに先立って特定の仕様の検査機器を用いて時系列的に検査データを取得し記録するデータ取得部として機能する（例えば、図８（ａ）、図８（ｂ）のＳ６１、Ｓ６３参照）。また、情報提供部１ｃは、推論の結果を伝達する情報伝達部として機能する。情報提供部１ｃは、推論部による推論結果から特定の人物用にカスタマイズした健康対応イベントを、特定の人物に伝達する情報伝達部として機能する。情報提供部１ｃは、特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得するデータ取得部として機能する。情報提供部１ｃは、データベースを用いて、変化パターンのトレンド変化とイベントのタイミングの関係に基づいて、イベント情報のうち、トレンド変化に影響したイベントを抽出して情報提供を可能とする情報提供部として機能する。 Further, the information providing unit 1c functions as a data acquisition unit that acquires inspection data in a time series of a specific period using an inspection device having a specific specification and acquires change pattern information of the inspection data (for example, FIG. 8 (b), see S61 and S63). The information providing unit 1c responds to the input of a specific health response event, and functions as a data acquisition unit that acquires and records inspection data in chronological order using an inspection device having specific specifications prior to the health response event ( For example, see S61 and S63 in FIGS. 8 (a) and 8 (b)). In addition, the information providing unit 1c functions as an information transmitting unit that transmits the result of inference. The information providing unit 1c functions as an information transmission unit that transmits a health response event customized for a specific person from the inference result by the inference unit to the specific person. The information providing unit 1c functions as a data acquisition unit that acquires a plurality of inspection data in a time series over a specific period using an inspection device having a specific specification. The information providing unit 1c uses a database to extract events that affect the trend change from the event information based on the relationship between the trend change of the change pattern and the timing of the event, and makes it possible to provide the information. Functions as.
　推論モデル仕様決定部１ｄは、推論依頼部１ｅが学習部５に推論モデルの生成を依頼する際に、生成する推論モデルの仕様を決定する。制御部１は、情報判定機器２等からユーザの生体情報を取得し、この生体情報を蓄積している。制御部１は、蓄積した生体情報を教師データとして、学習部５に種々の推論モデルの生成を依頼する。推論モデル仕様決定部１ｄは、推論モデルの生成に当たって、どのような仕様の推論モデルを依頼するかを決定する。例えば、後述する図３（ａ）に示すように、時系列的な生体情報が蓄積されている場合に、推論モデル仕様決定部１ｄは、どのような検査データ（値）となると、ユーザは何日後に医療施設で治療を受けることになるかを推論するための推論モデルの仕様を決定する。また、推論モデル仕様決定部１ｄは、時系列的な生体情報に基づいて、更に必要な検査や治療を受けるために推奨される施設を推論する推論モデルを生成するための仕様を決定する。 The inference model specification determination unit 1d determines the specifications of the inference model to be generated when the inference request unit 1e requests the learning unit 5 to generate the inference model. The control unit 1 acquires the biometric information of the user from the information determination device 2 and the like, and accumulates the biometric information. The control unit 1 requests the learning unit 5 to generate various inference models using the accumulated biological information as teacher data. The inference model specification determination unit 1d determines what kind of specification the inference model is requested in generating the inference model. For example, as shown in FIG. 3A, which will be described later, when time-series biometric information is accumulated, what kind of inspection data (value) does the inference model specification determination unit 1d have, and what is the user? Determine the specifications of an inference model to infer whether you will be treated in a medical facility in a day. In addition, the inference model specification determination unit 1d determines specifications for generating an inference model that infers a facility recommended for receiving further necessary tests and treatments based on time-series biological information.
　推論依頼部１ｅは、推論モデル仕様決定部１ｄによって決定された仕様の推論モデルの生成を、学習部５に依頼する。すなわち、推論依頼部１ｅは、情報判定機器２によって取得した生体情報が所定数蓄積している場合に、学習部５に推論モデルの生成を依頼し、生成された推論モデルを受信する。この受信した推論モデルは、推論エンジン７に送信される。なお、制御部１は、推論モデルを複数用意し、ユーザに提供すべき情報に応じて、適宜、推論モデルを選択するとよい。 The inference request unit 1e requests the learning unit 5 to generate an inference model of the specifications determined by the inference model specification determination unit 1d. That is, the inference requesting unit 1e requests the learning unit 5 to generate an inference model when a predetermined number of biological information acquired by the information determination device 2 is accumulated, and receives the generated inference model. This received inference model is transmitted to the inference engine 7. The control unit 1 may prepare a plurality of inference models and appropriately select the inference model according to the information to be provided to the user.
　検索部１ｆは、情報判定機器２によって取得されたユーザの生体情報に基づいて、さらに検査や治療が必要であることが判明した際に、検査や治療に必要な設備を有する検査機関や医療機関を、ＤＢ部８に蓄積されているデータベースの中で、検索を行う。検索部１ｆは、データベースを用いて、変化パターンのトレンド変化とイベントのタイミングの関係に基づいて、イベント情報のうち、トレンド変化に影響したイベントを検索する検索部として機能する。 The search unit 1f is an inspection institution or medical institution having equipment necessary for the inspection or treatment when it is found that further examination or treatment is necessary based on the biometric information of the user acquired by the information determination device 2. Is searched in the database stored in the DB unit 8. The search unit 1f functions as a search unit for searching for an event that affects the trend change in the event information based on the relationship between the trend change of the change pattern and the event timing using the database.
　情報判定機器２は、対象者の健康関連情報、例えば、バイタル情報、検体情報等の検査データを取得するための機器である。健康関連情報としては、種々の情報があり、例えば、対象者の体温、血圧、心拍等のバイタル情報がある。また健康関連情報としては、対象者の尿、大便等の排泄物や、痰や、血液等、種々の検体情報がある。大便の場合には、情報判定機器２は、その色、形状、量、日時情報を取得する。情報判定機器２は、制御部１からの指示に従って情報を取得してもよく、またユーザの操作に応じて情報を取得してもよく、また自動的に情報を取得してもよい。さらに、情報判定機器２は、医療・健康情報である情報「パーソナル・ヘルス・レコード（(Personal Health Records : PHR) 」に、日常生活、職場／学校での活動、食事、スポーツ活動など、日常生活の様々な活動データを加えたパーソナル・ライフ・レコード（Personal Life Records : PLR） を収集・活用してもよい。取得した情報は、情報判定機器２内の通信部（図示を省略）を通じて、制御部１に送信される。 The information determination device 2 is a device for acquiring test data such as health-related information of the target person, for example, vital information and sample information. As health-related information, there are various kinds of information, for example, vital information such as body temperature, blood pressure, and heartbeat of the subject. In addition, as health-related information, there are various sample information such as excrement such as urine and stool of the subject, sputum and blood. In the case of stool, the information determination device 2 acquires the color, shape, amount, and date / time information. The information determination device 2 may acquire information according to an instruction from the control unit 1, may acquire information according to a user's operation, or may automatically acquire information. Further, the information determination device 2 is used for daily life such as daily life, work / school activities, meals, sports activities, etc. in the information "Personal Health Records (PHR)" which is medical / health information. Personal Life Records (PLR), which includes various activity data of the above, may be collected and utilized. The acquired information is controlled through a communication unit (not shown) in the information judgment device 2. It is transmitted to part 1.
　情報判定機器２は、対象者の検査データを取得する検査データ取得部として機能する。また、情報判定機器２は、対象者の特定期間の時系列パターンとなる検査データを取得する検査データ取得部として機能する。検査データ取得部が取得する対象者の検査データは、特定の仕様の検査機器を用いて時系列的に検査データを取得し、該検査データの変化パターン情報を特定の時間幅で抽出したものである。すなわち、情報判定機器２として、特定仕様の検査機器（同一のタイプの検査機器）を用い、情報判定機器２が、同一の対象者の検査項目について、異なるタイミングで測定することによって、時系列的にデータを取得する。この時系列的なデータを用いて、検査タイミングに応じて測定値をグラフ上に描くことによって、変化パターンを得ることができる。この変化パターンを特定の時間幅で抽出することによって、検査データを得ることができる。検査データは、排便時用の色センサ、形状センサ、硬度センサ、嗅覚センサ（線虫や動物の反応判定を含む）、ガス成分センサ、特定の試薬添加時の色変化検出センサ、拡大観察画像による形状判定のいずれかの出力結果の一つに従って得られたデータである。 The information determination device 2 functions as an inspection data acquisition unit that acquires inspection data of the target person. In addition, the information determination device 2 functions as an inspection data acquisition unit that acquires inspection data that is a time-series pattern of the target person for a specific period. The inspection data of the target person acquired by the inspection data acquisition unit is obtained by acquiring inspection data in chronological order using an inspection device having a specific specification and extracting change pattern information of the inspection data in a specific time width. is there. That is, as the information judgment device 2, an inspection device of a specific specification (inspection device of the same type) is used, and the information judgment device 2 measures the inspection items of the same subject at different timings in chronological order. Get the data to. Using this time-series data, a change pattern can be obtained by drawing the measured values on a graph according to the inspection timing. Inspection data can be obtained by extracting this change pattern in a specific time width. The inspection data is based on a color sensor for defecation, a shape sensor, a hardness sensor, an olfactory sensor (including reaction judgment of nematodes and animals), a gas component sensor, a color change detection sensor when a specific reagent is added, and a magnified observation image. It is the data obtained according to one of the output results of any of the shape determinations.
　情報判定機器２としてウェアラブル端末を利用する場合には、ウェアラブル端末の装着部位によって、皮膚やあるいは身体近傍に密着し、体温、心拍、血圧、脳波、視線、呼吸、呼気などのバイタル情報を得ることが可能となる。また、体重計、血圧計、動脈壁の硬さを意味する動脈スティフネスを測定する測定器として、専用の精密な機器が、健康施設、公衆浴場、薬局、ショッピングモール等に配置され、さらに専門の計測者も一緒に配置されている場合がある。このような施設において、ユーザは空き時間などに測定機器を気楽に利用し、この時の測定結果に基づいて体調管理する場合も多い。これらの測定機器を情報判定機器２としてもよい。 When a wearable terminal is used as the information determination device 2, vital information such as body temperature, heartbeat, blood pressure, brain wave, line of sight, respiration, and exhalation can be obtained by closely contacting the skin or the vicinity of the body depending on the wearing part of the wearable terminal. Is possible. In addition, as a scale, a sphygmomanometer, and a measuring instrument for measuring arterial stiffness, which means the hardness of the arterial wall, dedicated precision equipment is installed in health facilities, public baths, pharmacies, shopping malls, etc. The measurer may also be assigned. In such facilities, users often use the measuring device comfortably in their spare time and manage their physical condition based on the measurement results at that time. These measuring devices may be used as the information determination device 2.
　また、情報判定機器２は、ユーザが専用の端末等を使用した前後に、アンケートに記入を依頼する場合がある。このような場合には、このアンケートの記載に基づいて、ユーザのプロフィール情報やその他の情報を特定できる。このような情報収集は、情報判定機器２に限らず、制御部１が行ってもよい。この情報は、後述する図４のステップＳ３における特定情報を取得したか否かの判定の際に使用することができる。何時、医者に行ったかの情報なども聞き取りできれば、後述する図３（ａ）（ｂ）における時刻Ｔｃ情報として使用することができる。 In addition, the information judgment device 2 may request the user to fill out a questionnaire before and after using a dedicated terminal or the like. In such cases, the user's profile information and other information can be identified based on the description in this questionnaire. Such information collection is not limited to the information determination device 2, and may be performed by the control unit 1. This information can be used when determining whether or not the specific information has been acquired in step S3 of FIG. 4, which will be described later. If information such as when the doctor went to the doctor can be heard, it can be used as the time Tc information in FIGS. 3 (a) and 3 (b) described later.
　情報判定機器２が、特定ユーザに関する情報を得た場合、制御部１の情報提供部１ｃが推奨する施設に関する情報を、特定ユーザの情報端末４に提示する。この提示が、ユーザの行動を補助することを想定して、説明を行うが、様々な変形が考えらる。 When the information determination device 2 obtains information on a specific user, the information on the facility recommended by the information providing unit 1c of the control unit 1 is presented to the information terminal 4 of the specific user. The explanation is given on the assumption that this presentation assists the user's behavior, but various variations can be considered.
　情報判定機器２は、すでに特定の疾患にかかっていて、医師の指導のもとで使用している体温計や血圧計などでもよい。また、スマートフォンの有するカメラで撮影した顔や爪などの色や顔の表情、患部の画像、喉がおかしくなった時の声をマイクで収音する場合等では、携帯端末（スマートフォン）がそのまま情報判定機器２となりうる。 The information determination device 2 may be a thermometer or a sphygmomanometer that is already suffering from a specific disease and is used under the guidance of a doctor. In addition, when the color and facial expression of the face and nails taken by the camera of the smartphone, the image of the affected area, and the voice when the throat becomes strange are picked up by the microphone, the mobile terminal (smartphone) provides the information as it is. It can be the judgment device 2.
　最近では、簡易の健康管理機器や健康情報取得機器が開発されており、これらの機器がウェアラブル機器に搭載される場合がある、このような装置もスタンドアローンではなく、スマートフォンの周辺機器として扱われる場合が多いので、これも携帯端末として想定してもよい。また、ウェアラブルな機器でなくとも、簡易な測定機器を、人が集まる場所に設置し、健康情報サービスを提供している場合がある。このような機器を情報判定機器２として利用してもよい。 Recently, simple health management devices and health information acquisition devices have been developed, and these devices may be installed in wearable devices. Such devices are also treated as peripheral devices for smartphones, not stand-alone devices. Since there are many cases, this may also be assumed as a mobile terminal. In addition, even if it is not a wearable device, a simple measuring device may be installed in a place where people gather to provide a health information service. Such a device may be used as the information determination device 2.
　情報判定機器２等において行われる情報判定は、どこまで判定するかは、制御部１との関係で変更してもよい。例えば、情報判定機器２等においてセンシングした結果のみを判定せずに制御部１に送信してもよい。ただし、この場合には、どのような人のどのようなデータであるかの情報を、センシング信号に添付し、この信号を送信する必要がある。この添付情報は、どの人か、どのセンシング結果か、が対応付けられていることが好ましいが、別の端末の情報に、加味することによって、対応付けてもよい。 The extent to which the information determination performed by the information determination device 2 or the like is determined may be changed in relation to the control unit 1. For example, the information determination device 2 or the like may transmit the sensing result to the control unit 1 without determining only the result. However, in this case, it is necessary to attach information on what kind of data of what kind of person to the sensing signal and transmit this signal. It is preferable that the attached information is associated with which person and which sensing result, but it may be associated with the information of another terminal by adding it.
　関連検査機関９は、ユーザが検査を受ける施設であり、例えば、検査施設や医療施設がある。この関連検査機関９は、移動型、例えば、自動車、列車、船、ヘリコプター、ドローン等に一般医療機器や検査機器を搭載し、患者のもとに出向くタイプであっても勿論構わない。制御部１は、どの医療機関に行ったか、またどのような検査結果が出たかなどを関連検査機関９のシステムを運営するサーバなどから取得可能である。もちろん、関連検査機関９のサーバが制御部１と同じであってもよく、また一部の機能を分担してもよい。 The related inspection institution 9 is a facility where the user is inspected, and there are, for example, an inspection facility and a medical facility. Of course, the related inspection organization 9 may be a mobile type, for example, a type in which general medical equipment or inspection equipment is mounted on an automobile, train, ship, helicopter, drone, or the like and the patient goes to the patient. The control unit 1 can acquire which medical institution the patient went to, what kind of test result was obtained, and the like from the server that operates the system of the related test institution 9. Of course, the server of the related inspection organization 9 may be the same as that of the control unit 1, or some functions may be shared.
　関連検査機関９として、上述の説明では、具体的な生物学的な生体情報や生体からの試料サンプルを用いた医療関連の検査を行うものを記載した。しかし、問診などや自覚症状などの質疑応答なども検査であると考えれば、その問診や質疑応答をもとに自然言語的な入力によって、何らかの結果を出すものも検査機関の一つだと考えられる。つまり、アンケート記載または項目選択によって、特定の健康アドバイスを行うインターネット・サービスなどもこの関連検査機関９を広義に解釈したものに含まれると考えてもよい。この広義の関連検査機関９は、表示された画面の特定の領域ごとに整理されて紐づけられたカテゴリーに対応した情報を入力した場合に、項目の内容毎にデータベース検索やロジックベースの分岐や推論処理によって、健康関係アドバイスを文字化して表示し、あるいは音声化して音声でユーザに伝える技術を使用したサービスを提供することができる。 As the related testing institution 9, in the above description, a medical-related test using specific biological biological information or a sample sample from a living body is described. However, if we consider that questions and answers such as interviews and subjective symptoms are also tests, we think that one of the inspection institutions is one that produces some results by inputting in natural language based on the questions and answers. Be done. In other words, it may be considered that the Internet service that provides specific health advice by questionnaire description or item selection is also included in the broad interpretation of the related inspection organization 9. When the related inspection organization 9 in this broad sense inputs information corresponding to the category organized and linked for each specific area of the displayed screen, database search and logic-based branching are performed for each item content. By inference processing, it is possible to provide a service using a technique of displaying health-related advice in characters or by voicing it to the user by voice.
　端末４は、携帯情報端末であり、対象者やその関係者が確認可能な情報を受け取るための装置である。端末４は、例えばスマートフォンやタブレットＰＣであってもよく、この場合には、内蔵カメラやマイクを情報取得部として利用することができる。また、連携可能なウェアラブル端末その他の家電を端末４として使用してもよく、ウェアラブル端末等によって情報を取得してもよい。したがって、情報判定機器２と端末４は同じものであってもよく、またそれぞれ専用機器であってもよい。さらに、状況に応じて、制御部１が有する機能を情報判定機器２や端末４が有してもよく、分担して検出や制御や情報提供を行うような構成にしてもよい。 The terminal 4 is a mobile information terminal, and is a device for receiving information that can be confirmed by the target person and related persons. The terminal 4 may be, for example, a smartphone or a tablet PC. In this case, the built-in camera or microphone can be used as an information acquisition unit. Further, a wearable terminal or other home appliances that can be linked may be used as the terminal 4, and information may be acquired by the wearable terminal or the like. Therefore, the information determination device 2 and the terminal 4 may be the same device, or may be dedicated devices, respectively. Further, depending on the situation, the information determination device 2 or the terminal 4 may have the function of the control unit 1, and may be configured to share detection, control, and information provision.
　データベース（ＤＢ）部８は、医療施設等が保有する機器に関する情報のデータベースである。保有機器としては、診断を行うために使用する測定機器や、治療を行うための器具・機器等がある。推奨する施設に関する情報の提供は、制御部１がデータベース部８に記録された情報を参照して行う。 The database (DB) section 8 is a database of information on devices owned by medical facilities and the like. Owned equipment includes measuring equipment used for diagnosis, equipment / equipment for performing treatment, and the like. The control unit 1 refers to the information recorded in the database unit 8 to provide the information regarding the recommended facility.
　ＤＢ部８は、電気的に書き換え可能な不揮発性メモリを有する。ＤＢ部８は、施設別保有機器一覧を記録した記録部８ａと、施設別にＩＤとそのユーザの来院を記録した記録部８ｂを有している。記録部８ａは、病院やクリニックや検査機関など施設が保有する機器の一覧を記録する。情報提供部１ｃは、記録部８ａを検索することによって、検査に最適な機器がある施設の情報をユーザに提示することが可能となる。医療施設等が装置を買い替えた場合等に応じて、情報をアップデートするために、関連検査機関９の情報と連携してもよい。また、記録部８ｂは、施設ごとに、どの人（ＩＤで特定される）が何時来たかという来院情報を記録する。記録部８ｂは省略してもよい。 The DB unit 8 has an electrically rewritable non-volatile memory. The DB unit 8 has a recording unit 8a that records a list of owned devices for each facility, and a recording unit 8b that records an ID and a user's visit for each facility. The recording unit 8a records a list of devices owned by facilities such as hospitals, clinics, and inspection institutions. By searching the recording unit 8a, the information providing unit 1c can present the information of the facility having the most suitable equipment for the inspection to the user. In order to update the information when the medical facility or the like replaces the device, the information may be linked with the information of the related inspection institution 9. In addition, the recording unit 8b records visit information indicating which person (identified by ID) came at what time for each facility. The recording unit 8b may be omitted.
　ＤＢ部８は、医療施設が連携する情報伝達システムの一部に構築されており、ＤＢ部８が制御部１を通じて、関連検査機関９についてもアクセス可能としてもよい。この場合には、制御部１からＤＢ部８が検索命令を受けると、ＤＢ部８がＤＢ部８内に記録されたデータに加えて、関連検査機関９内のデータについても検索を行い、検索結果を出力する。ＤＢ部８は、対象者のプロフィール情報と、検査・医療機関ごとの保有機器情報を記憶する記憶部として機能する。なお、この記憶部は、ＤＢ部８に限らず、制御部１内等にその機能の全部、または一部を配置してもよい。ＤＢ部８において記録されるデータの詳細については、図２を用いて後述する。 The DB unit 8 is constructed as a part of an information transmission system in which medical facilities cooperate, and the DB unit 8 may also be able to access the related inspection institution 9 through the control unit 1. In this case, when the DB unit 8 receives a search command from the control unit 1, the DB unit 8 searches the data in the related inspection organization 9 in addition to the data recorded in the DB unit 8 to search. Output the result. The DB unit 8 functions as a storage unit that stores the profile information of the target person and the possessed device information for each examination / medical institution. The storage unit is not limited to the DB unit 8, and all or part of its functions may be arranged in the control unit 1 or the like. Details of the data recorded in the DB unit 8 will be described later with reference to FIG.
　ＤＢ部８は、変化パターン情報に対し、服薬や生活改善の開始など特定の改善タイミング情報を変化パターン情報の時間情報と同様の時間情報として含むことによって、この改善情報以前の変化パターンの部分を比較可能にした情報を有するデータベースを作成するデータ作成部として機能する。ＤＢ部８は、検査データを取得して作成した検査データの変化パターン情報と、特定の期間内における検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録可能なデータベースを作成するデータ作成部として機能する。 The DB unit 8 includes the change pattern information with specific improvement timing information such as taking medication and the start of life improvement as time information similar to the time information of the change pattern information, so that the part of the change pattern before the improvement information is included. It functions as a data creation unit that creates a database with comparable information. The DB unit 8 creates a database that can record the change pattern information of the inspection data created by acquiring the inspection data and the timing of each of the plurality of event information regarding the acquisition source of the inspection data within a specific period in association with each other. It functions as a data creation unit to be created.
　学習部５は、入出力モデル化部５ａを有し、機械学習等によって推論モデルを生成する。この推論モデルは、取得した生体情報、生検情報など取得情報と疾患の関係を学習し、具体的には、取得情報と診療科・部門の関係を学習することによって生成する。入出力モデル化部５ａは、推論エンジン７と同様に、入力層、複数の中間層、出力層を有し、中間層のニューロンの結合の強さを学習によって求め、推論モデルを生成する。 The learning unit 5 has an input / output modeling unit 5a and generates an inference model by machine learning or the like. This inference model is generated by learning the relationship between acquired information such as acquired biometric information and biopsy information and the disease, and specifically by learning the relationship between the acquired information and the clinical department / department. Similar to the inference engine 7, the input / output modeling unit 5a has an input layer, a plurality of intermediate layers, and an output layer, obtains the strength of the connection of neurons in the intermediate layer by learning, and generates an inference model.
　学習部５は、変化パターン情報に対し、特定期間内で上記変化パターンが変化した時点に対応する時点における影響イベント情報を検出し、検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する学習部として機能する（例えば、図８のＳ６５～Ｓ７１参照）。また、学習部５は、記録されていた検査データについての変化パターン情報に対し、健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する学習部として機能する（例えば、図８のＳ６５～Ｓ７１参照）。 The learning unit 5 detects the influence event information at the time corresponding to the time when the change pattern changes within a specific period with respect to the change pattern information, and learns the result of annotating the detected influence event information as teacher data. It functions as a learning unit for acquiring the inference model (see, for example, S65 to S71 in FIG. 8). In addition, the learning unit 5 annotates the change pattern information of the recorded test data with information on the equipment, / or equipment, and / or environment at the time of the health response event as teacher data. It functions as a learning unit that acquires an inference model trained as (see, for example, S65 to S71 in FIG. 8).
　このような推論モデルの生成にあたっては、同一タイプの検査機器（特定の仕様の検査機器）を用いて被検者から取得した検査データの変化パターンを特定の時間幅で抽出し、この抽出した変化パターンを推論エンジン７に入力し、被検者が検査したタイミングから、後のタイミングにおいて出力されるべき健康アドバイスをアノテーション情報とした、教師データを生成する。そして、この教師データを用いて学習を行うことによって、推論モデルが生成される。なお、本実施形態においては、検査結果が出た時点から遡る時間幅について説明したが、検査結果の後、治療等でデータが良くなる場合、治療がうまくいっていない場合があるので、その差異を学習して、予後のアドバイスを出力してもよい。 In generating such an inference model, the change pattern of the test data acquired from the subject using the same type of test device (test device with specific specifications) is extracted in a specific time width, and the extracted change is extracted. The pattern is input to the inference engine 7, and the teacher data is generated using the health advice to be output at a later timing as the annotation information from the timing of the examination by the subject. Then, an inference model is generated by performing learning using this teacher data. In this embodiment, the time width that goes back from the time when the test result is obtained has been described, but if the data improves after the test result due to treatment or the like, the treatment may not be successful. You may learn and output prognostic advice.
　また、学習部５は、検査、通院、服薬の後の検査データ列を用いて学習すれば、生活習慣改善や治療や服薬の効果の将来予想アドバイスを行うことが可能な推論モデルを生成することも出来る。この場合には、検査、通院、服薬の時点を起点として、その後の時系列データを利用する。検査、通院、服薬などをアドバイスする場合は、この前の時系列データを利用する。 In addition, the learning unit 5 can generate an inference model capable of providing lifestyle-related improvement and future prediction advice on the effects of treatment and medication by learning using the examination data sequence after examination, hospital visit, and medication. You can also do it. In this case, the time-series data is used starting from the time of examination, outpatient visit, and medication. Use the previous time-series data when giving advice on tests, hospital visits, medications, etc.
　ここで、学習部５が行う学習の一例として、深層学習について、説明する。「深層学習（ディープ・ラーニング）」は、ニューラル・ネットワークを用いた「機械学習」の過程を多層構造化したものである。情報を前から後ろに送って判定を行う「順伝搬型ニューラル・ネットワーク」が代表的なものである。順伝搬型ニューラル・ネットワークは、最も単純なものでは、Ｎ１個のニューロンで構成される入力層、パラメータで与えられるＮ２個のニューロンで構成される中間層、判別するクラスの数に対応するＮ３個のニューロンで構成される出力層の３層があればよい。入力層と中間層、中間層と出力層の各ニューロンはそれぞれが結合加重で結ばれ、中間層と出力層はバイアス値が加えられることによって、論理ゲートを容易に形成できる。 Here, deep learning will be described as an example of learning performed by the learning unit 5. "Deep learning" is a multi-layered structure of the process of "machine learning" using a neural network. A typical example is a "forward propagation neural network" that sends information from front to back to make a judgment. The simplest forward-propagating neural network is an input layer consisting of N1 neurons, an intermediate layer consisting of N2 neurons given by parameters, and N3 corresponding to the number of classes to be discriminated. It suffices if there are three layers of output layers composed of the above neurons. Each neuron in the input layer and the intermediate layer, and each neuron in the intermediate layer and the output layer is connected by a connection weight, and a logic gate can be easily formed in the intermediate layer and the output layer by applying a bias value.
　ニューラル・ネットワークは、簡単な判別を行うのであれば３層でもよいが、中間層を多数にすることによって、機械学習の過程において複数の特徴量の組み合わせ方を学習することも可能となる。近年では、９層～１５２層のものが、学習にかかる時間や判定精度、消費エネルギーの観点から実用的になっている。また、画像の特徴量を圧縮する、「畳み込み」と呼ばれる処理を行い、最小限の処理で動作し、パターン認識に強い「畳み込み型ニューラル・ネットワーク」を利用してもよい。また、より複雑な情報を扱え、順番や順序によって意味合いが変わる情報分析に対応して、情報を双方向に流れる「再帰型ニューラル・ネットワーク」（全結合リカレントニューラルネット）を利用してもよい。 The neural network may have three layers as long as it makes a simple discrimination, but by increasing the number of intermediate layers, it is possible to learn how to combine a plurality of features in the process of machine learning. In recent years, those having 9 to 152 layers have become practical from the viewpoints of learning time, determination accuracy, and energy consumption. Alternatively, a "convolutional neural network" that compresses the feature amount of the image, performs a process called "convolution", operates with the minimum processing, and is strong in pattern recognition may be used. In addition, a "recurrent neural network" (fully connected recurrent neural network) that can handle more complicated information and whose meaning changes depending on the order or order may be used.
　これらの技術を実現するために、ＣＰＵやＦＰＧＡ（Field Programmable Gate Array）等の従来からある汎用的な演算処理回路を使用してもよい。しかし、これに限らず、ニューラル・ネットワークの処理の多くが行列の掛け算であることから、行列計算に特化したＧＰＵ（Graphic Processing Unit）やTensor Processing Unit（TPU）と呼ばれるプロセッサを利用してもよい。近年ではこのような人工知能（ＡＩ）専用ハードの「ニューラル・ネットワーク・プロセッシング・ユニット（ＮＰＵ）」がＣＰＵ等その他の回路とともに集積して組み込み可能に設計され、処理回路の一部になっている場合もある。 In order to realize these technologies, a conventional general-purpose arithmetic processing circuit such as a CPU or FPGA (Field Programmable Gate Array) may be used. However, not limited to this, since most of the processing of neural networks is matrix multiplication, even if a processor called GPU (Graphic Processing Unit) or Tensor Processing Unit (TPU) specialized in matrix calculation is used. Good. In recent years, such a "neural network processing unit (NPU)" dedicated to artificial intelligence (AI) has been designed so that it can be integrated and incorporated together with other circuits such as a CPU, and has become a part of processing circuits. In some cases.
　その他、機械学習の方法としては、例えば、サポートベクトルマシン、サポートベクトル回帰という手法もある。ここでの学習は、識別器の重み、フィルター係数、オフセットを算出するものあり、これ以外にも、ロジスティック回帰処理を利用する手法もある。機械に何かを判定させる場合、人間が機械に判定の仕方を教える必要がある。本実施形態においては、画像の判定を、機械学習によって導出する手法を採用したが、そのほか、人間が経験則・ヒューリスティクスによって獲得したルールを適応するルールベースの手法を用いてもよい。 Other methods of machine learning include, for example, support vector machines and support vector regression. The learning here is to calculate the weight of the discriminator, the filter coefficient, and the offset, and there is also a method using logistic regression processing. When making a machine judge something, humans need to teach the machine how to make a judgment. In the present embodiment, a method of deriving the judgment of the image by machine learning is adopted, but in addition, a rule-based method of applying the rules acquired by humans by empirical rules / heuristics may be used.
　推論エンジン７は、学習部５の入出力モデル化部５ａと同様の入出力層、ニューラル・ネットワークを有している。推論エンジン７は、学習部５によって生成された推論モデルを用いて、推論を行う。例えば、推論エンジン７は、情報判定機器２によって測定され、時系列的な生体情報を入力し、例えば、ユーザの健康状態を検査、治療等を行うに適切な検査機関・医療機関を推論によって求める。また、時系列的な生体情報に基づいて、いつ頃、医療機関で受診を受けることになるかの推論等を行ってもよい。 The inference engine 7 has an input / output layer and a neural network similar to the input / output modeling unit 5a of the learning unit 5. The inference engine 7 makes inferences using the inference model generated by the learning unit 5. For example, the inference engine 7 is measured by the information determination device 2, inputs time-series biometric information, and, for example, obtains an appropriate inspection institution / medical institution for inspecting, treating, or the like of the user's health condition by inference. .. In addition, based on time-series biometric information, it may be inferred when a medical institution will receive a medical examination.
　推論エンジン７は、対象者の検査・医療機関への来院のタイミング情報に従って学習された推論モデルを有する推論部として機能する。推論エンジン７は、対象者の検査データと同様の検査データを提供した者の変化パターンと、検査・医療機関への来院・検査・服薬情報に従って学習された推論モデルを有する推論部として機能する。推論部は、特定の仕様の検査機器を用いて対象者以外の者の検査データを取得し、この検査データの変化パターンを特定の時間幅で抽出したものを学習用推論部の入力とし、被検者が測定を行った時間幅の終了時点から後の時点において出力すべき被検者のための健康アドバイスをアノテーション情報として、学習用推論部が学習することによって生成した推論モデルを用いて推論を行う。 The inference engine 7 functions as an inference unit having an inference model learned according to the timing information of the subject's examination / visit to the medical institution. The inference engine 7 functions as an inference unit having an inference model learned according to the change pattern of the person who provided the examination data similar to the examination data of the subject and the examination / visit to a medical institution / examination / medication information. The inference unit acquires inspection data of a person other than the target person using an inspection device of a specific specification, extracts the change pattern of the inspection data in a specific time width, and uses it as an input of the inference unit for learning. Inference using the inference model generated by the learning inference unit using the health advice for the subject to be output from the end of the time width measured by the examiner to a later time as annotation information. I do.
　推論エンジン７は、特定の人物の上記特定の仕様の検査機器を用いて上記特定期間の幅と類似の期間で時系列的に得た検査データを、上記推論モデルに入力して影響イベントを推論する推論部として機能する（例えば、図８（ｂ）において生成した推論モデルを用いて、図４、図７等の検査結果の送信の際に推論する）。推論エンジン７は、特定の人物の特定の仕様の検査機器を用いて時系列的に得た検査データを、推論モデルに入力して設備、および／または備品、および／または環境の情報を推論する推論部として機能する（例えば、図８（ｂ）において生成した推論モデルを用いて、図４、図７等の検査結果の送信の際に推論する）。 The inference engine 7 inputs the inspection data obtained in time series in a period similar to the width of the specific period using the inspection device of the specific specification of the specific person into the inference model to infer the influence event. It functions as an inference unit (for example, inferring when transmitting the inspection results of FIGS. 4, 7, etc. using the inference model generated in FIG. 8 (b)). The inference engine 7 inputs inspection data obtained in time series using an inspection device of a specific specification of a specific person into an inference model to infer information on equipment and / or equipment and / or environment. It functions as an inference unit (for example, using the inference model generated in FIG. 8 (b), inference is made when transmitting the inspection results of FIGS. 4 and 7).
　また、対象者の来院・検査・服薬情報に従って学習された推論モデルは、入力を時系列の検査データとし、出力を関連疾病情報としてアノテーションされた教師データによって、入出力関係が設定された推論モデルである。ここで、関連疾病情報は、来院日時または診断結果、処方薬情報等である。 In addition, the inference model learned according to the subject's visit / examination / medication information is an inference model in which the input / output relationship is set by the teacher data annotated with the input as time-series examination data and the output as related disease information. Is. Here, the related disease information is the date and time of the visit, the diagnosis result, the prescription drug information, and the like.
　このように、制御部１は、検索部１ｆがＤＢ部８を検索する以外にも、推論エンジン７を利用して、推奨施設に関する情報を提供しても良い。推論エンジン７は、学習部５が生成した推論モデルを用いて、推奨施設に関する情報の推論を行う。この推論モデルは、取得した生体情報、生検情報など取得情報と疾患の関係を学習し、具体的には、取得情報と診療科・部門の関係を学習することによって生成する。このように、制御部１は、推論エンジン７による推論によっても、提示すべきガイド情報を出力してもよい。 As described above, the control unit 1 may provide information on the recommended facility by using the inference engine 7 in addition to the search unit 1f searching the DB unit 8. The inference engine 7 infers information about the recommended facility using the inference model generated by the learning unit 5. This inference model is generated by learning the relationship between acquired information such as acquired biometric information and biopsy information and the disease, and specifically by learning the relationship between the acquired information and the clinical department / department. In this way, the control unit 1 may output the guide information to be presented by the inference by the inference engine 7.
　制御部１が検索によって、または推論によって、一度に得られた取得情報に基づいて、一回の判定で医療施設などをガイドすると、いたずらに生活に医療情報を持ち込んで、健全に安心して生活するのを妨げる可能性がある。そこで、複数回の取得情報の履歴（時系列的情報）を用いて、精度アップしてもよい。 When the control unit 1 guides medical facilities, etc. with a single judgment based on the acquired information obtained at one time by search or inference, it unnecessarily brings medical information into life and lives soundly and with peace of mind. May interfere with. Therefore, the accuracy may be improved by using the history (time-series information) of the acquired information a plurality of times.
　図３（ａ）に、記録部８に記録された個人の健康関連の履歴データ（時系列的データ）を用いたグラフを示す。この記録部８には、例えば、特定機器Ａによって取得されたデータ、または様々の検査機能がある機器のデータの内、特定のデータＡと、その個人が何時、どの施設を受診したかを記録している。制御部１は記録部８の記録を管理している。 FIG. 3A shows a graph using personal health-related historical data (time-series data) recorded in the recording unit 8. The recording unit 8 records, for example, the specific data A among the data acquired by the specific device A or the data of the device having various inspection functions, and when and which facility the individual visited. doing. The control unit 1 manages the recording of the recording unit 8.
　図３（ａ）に示すグラフの横軸は時間であり、縦軸は健康関連データである。このグラフ上に転記したデータはあたかも二次元上に情報を配置した画像データとして扱うことができる。そこで、この画像から特定のものを見つけ出すという、画像検索と同様の手法によって、推論が可能である。つまり、グラフが入力となり出力を医療機関情報とすればよい。図３に示す履歴データを示すグラフの詳細については後述する。 The horizontal axis of the graph shown in FIG. 3 (a) is time, and the vertical axis is health-related data. The data posted on this graph can be treated as if the information were arranged two-dimensionally. Therefore, inference can be made by a method similar to image search, in which a specific object is found from this image. That is, the graph may be the input and the output may be the medical institution information. Details of the graph showing the historical data shown in FIG. 3 will be described later.
　記録部８に、健康関係情報が変化した人が、どのような施設に行くことになったか、あるいは行っているかを記録し、制御部１がこの記録を一元管理すると、これらのデータを集め、教師データとして用いて、学習部５に推論モデルを作成させることが可能となる。 The recording unit 8 records what kind of facility the person whose health-related information has changed is going to or is going to, and when the control unit 1 centrally manages this record, these data are collected. It is possible to have the learning unit 5 create an inference model by using it as teacher data.
　推論エンジン７は、制御部１の推論モデル仕様決定部１ｄが推論モデルの仕様を指定し、この仕様に従った学習によって得られた推論モデルを有する。推論エンジン７は、新たな機器が登場した場合には、機器データが異なる場合があるので、推論モデルは、複数あってもよい。また、新たな機器が登場するたびに、新たな学習が必要になることから、制御部１の指定で学習部５を通じて、推論モデルが改良されたり新作されたりすることが多いことを想定している。ただし、情報判定機器２が専用となっており、特定の疾患の推論のみに特化した場合は、単独の専用推論モデルであってもよい。 The inference engine 7 has an inference model obtained by inference model specification determination unit 1d of the control unit 1 designating an inference model specification and learning according to this specification. The inference engine 7 may have a plurality of inference models because the device data may be different when a new device appears. In addition, since new learning is required each time a new device appears, it is assumed that the inference model is often improved or newly created through the learning unit 5 by the designation of the control unit 1. There is. However, when the information determination device 2 is dedicated and specialized only in inference of a specific disease, it may be a single dedicated inference model.
　なお、推論エンジン７は、学習部５の入出力部５ａと同様に、ＣＰＵ、ＧＰＵ、ＤＳＰなどＡＩチップを中心とした回路ブロックで、メモリなども搭載してニューラル・ネットワークを構成している。この推論エンジン７や学習部５が、病院などが連携して運営するネットワーク等に接続されており、これらと協働して制御部１が利用できる場合があり得ることを、本実施形態においては想定している。この場合、関連検査機関９を経由して、学習や推論の情報がやり取りできる可能性もある。 The inference engine 7 is a circuit block centered on an AI chip such as a CPU, GPU, and DSP, like the input / output unit 5a of the learning unit 5, and also includes a memory and the like to form a neural network. In the present embodiment, the inference engine 7 and the learning unit 5 are connected to a network or the like operated in cooperation with a hospital or the like, and the control unit 1 may be used in cooperation with these. I'm assuming. In this case, there is a possibility that learning and inference information can be exchanged via the related inspection organization 9.
　制御部１の推論依頼部１ｅが情報判定機器２等から特定のユーザの情報を十分取得できたと判定した時に、推論エンジン７に推論を依頼する。推論エンジン７は、類似のデータの時系列的な推移を示す情報群を入力し、この情報群に基づいて推論し、特定のユーザに適した医療機関情報（来院情報、検査情報など）を出力することが可能である。常に通院しており、慢性化した病状を有する人と類似のデータ群が推論エンジン７に入力された場合は、同様の治療が出来る施設がガイド表示された方が良い。推論エンジン７は、特定のユーザを検査すべき機器を保有している医療機関の推定が可能である。 When the inference request unit 1e of the control unit 1 determines that the information of a specific user has been sufficiently acquired from the information determination device 2 or the like, the inference engine 7 is requested to make an inference. The inference engine 7 inputs an information group showing the time-series transition of similar data, infers based on this information group, and outputs medical institution information (visit information, examination information, etc.) suitable for a specific user. It is possible to do. If a group of data similar to a person with a chronic medical condition is input to the inference engine 7, it is better to display a guide to a facility where the same treatment can be performed. The inference engine 7 can estimate the medical institution that has the equipment to inspect a specific user.
　また、今後、家電等の機器に健康見守りの機能が搭載される可能性は高まるので、ユーザが特別な設備が設置された場所に行かずとも、これらの機器によって生活の中で様々な情報が取得され、ユーザが意識せずとも有効な健康管理が出来る。例えば、温水便座や便器などに取り付けたセンサが、便の量や色を検知し、この検知結果を診断に生かす方法などが多く提案されている。 In addition, since it is more likely that home appliances and other devices will be equipped with health monitoring functions in the future, these devices will provide various information in daily life without the user having to go to a place where special equipment is installed. It is acquired and effective health management can be performed without the user being aware of it. For example, many methods have been proposed in which a sensor attached to a warm water toilet seat or a toilet bowl detects the amount and color of stool, and the detection result is used for diagnosis.
　次に、図２を用いて、ＤＢ部８に記録されているデータの一例を説明する。図２の表は、情報判定機器２が出力するデータ（検査データ、生体データ、バイタルデータ、検体データなどを総じて「時系列データ」として表記）がどのような時間的変化、推移となった患者（患者ＩＤとして説明）が、どのような医療機関等と関連したかを示す。この表から、患者の行動履歴や院内システムの来歴に基づいて、患者が、何れかの病院や診療科に、何時来たかなどが分かるようになっている。 Next, an example of the data recorded in the DB unit 8 will be described with reference to FIG. The table of FIG. 2 shows the temporal changes and transitions of the data output by the information determination device 2 (test data, biological data, vital data, sample data, etc. are generally expressed as "time series data"). (Explanation as patient ID) indicates what kind of medical institution, etc. was associated with. From this table, it is possible to know when the patient came to any hospital or clinical department based on the patient's behavior history and the history of the in-hospital system.
　ＤＢ部８に記録されている病院のデータとしては、どのような診療科があり、どのような疾病の治療に実績があるか、またどのような医師、看護師その他スタッフが勤務しており、どのような検査が受けられるか、またどれくらいの待ち時間であるか、また予約制か紹介制かを含めて時間割等の受診条件等の情報が整理されて記録されている。受診条件には、どのような問診票を記入しなければならないかなどの情報を含んでもよい。また、受診条件には、受診や通院、手術関係の情報のみならず、関連リハビリ施設の情報、予後の通院、処置情報なども含まれていてもよい。 The hospital data recorded in the DB department 8 includes what kind of clinical departments there are, what kind of illnesses have been treated, and what kind of doctors, nurses and other staff are working. Information such as the timetable and other consultation conditions, including what kind of examination can be taken, how long the waiting time is, and whether it is a reservation system or a referral system, is organized and recorded. The consultation conditions may include information such as what kind of questionnaire must be filled out. In addition, the consultation conditions may include not only information related to consultation, hospital visit, and surgery, but also information on related rehabilitation facilities, prognosis visit, treatment information, and the like.
　また、データを表形式によって記録することは、病院に限らず、薬局や公共施設で検査を受ける場合でも対応が可能である。薬局に関するデータとしては、どのような医薬品が準備され、提供されているかなどの情報があってもよい。 In addition, recording data in a tabular format is possible not only in hospitals but also in pharmacies and public facilities when undergoing examinations. As data on pharmacies, there may be information such as what kind of medicines are prepared and provided.
　医療機関情報として、上述した情報や、併設された薬局の特色も含めて整理されていれば、特定の病状の治療に特化した医療機関にこだわらず、近所で足を運びやすい類似の特色を有する医療機関に行くことで代用が出来る。単に、各病院・クリニックが専門としていたり、或いは有していたりする診療科といった特徴を超えて、疾病の特徴に応じて適切な医療機関を選んで訪問することが可能となる。このような情報は、勤務する医療従事者の異動、退職など、さらには機器の買い替え等があることから、適当な周期にて見直しが必要である。 If the medical institution information is organized including the above-mentioned information and the characteristics of the attached pharmacy, it will be easy to visit in the neighborhood regardless of the medical institution specializing in the treatment of a specific medical condition. You can substitute it by going to your own medical institution. It is possible to select and visit an appropriate medical institution according to the characteristics of the disease, beyond the characteristics of the clinical departments that each hospital / clinic specializes in or has. Such information needs to be reviewed at an appropriate cycle because there are changes in medical staff who work, retirement, and replacement of equipment.
　したがって、医療機関情報に変化がある都度、あるいは定期的な状況変化を反映させるデータベースである必要がある。上述したような工夫を行うことによって、単に病名とか自覚症状のみならず、情報判定機器２が出力する患者のデータ（検査データ、生体データ、バイタルデータ、検体データなどを総じて「時系列データ」として表記）が、どのような時間的変化、推移となったかに基づいて、単に特定の診療科というカテゴリー以上に、細かく分類整理された特徴を持つ医療機関における受診と、この受診よって快方に向かったかどうかという関係性を把握することができる。このため、これまで以上にカスタマイズされた情報伝達装置および情報伝達方法が提供可能となる。 Therefore, it is necessary to have a database that reflects changes in medical institution information each time or on a regular basis. By devising as described above, not only the disease name and subjective symptoms but also the patient data (examination data, biological data, vital data, sample data, etc.) output by the information determination device 2 are collectively regarded as "time series data". Based on what kind of temporal change and transition the notation) has changed, the consultation at a medical institution that has characteristics that are more finely classified and organized than the category of a specific clinical department, and this consultation will improve the situation. It is possible to grasp the relationship of whether or not it was. Therefore, it becomes possible to provide a more customized information transmission device and information transmission method than ever before.
　また、医療機関情報としては、受診、通院等に限らず、未病、予後の情報があってもよく、患者が訪れた健康関連施設（健康を増進させ、或いは回復を促す公共機関・施設等、或いは摂取することによって体調に影響する飲食物を提供する店舗等の情報があってもよい）にどのような設備があるかも一覧で表示している。大きな病院では多くの設備があり、小さなクリニックには限られた施設しかない。その疾病に関する症例数も大きな病院の方が多い状況になりがちである。しかし、特定の疾病に特化した専門のクリニックが、小さなクリニックを取り巻く環境を補い、スマートシティ等においてクリニックの連携によってあたかも大きな病院と同様の機能を有することができる。そこで、ここでは、有効な情報として、患者に提示することもできるようにしている。 In addition, medical institution information is not limited to consultations and outpatient visits, but may include information on non-illness and prognosis, and health-related facilities visited by patients (public institutions / facilities that promote or promote recovery of health, etc.) Or, there may be information on stores that provide food and drink that affects the physical condition by ingestion), and what kind of equipment is available is also displayed in a list. Large hospitals have many facilities, and small clinics have limited facilities. The number of cases related to the disease tends to be higher in large hospitals. However, a specialized clinic specializing in a specific disease can supplement the environment surrounding a small clinic, and can have the same function as a large hospital by cooperating with the clinic in a smart city or the like. Therefore, here, it is possible to present it to the patient as effective information.
　近年、コンビニエンスストアなどに様々な情報入出力機能が充実しており、これらの店舗を中心に、特定個人の購入履歴や入金履歴の特徴を割り出して、個々人の有する特徴的なニーズを満たすサービスが構築されている。そこで、これらの情報と連携して、上述の情報提供を、購買行動の中に反映させて行ってもよい。例えば、血圧が高くなってきている人には、高血圧症状を防止するような購買アドバイス、あるいは食事アドバイスを店舗において行ってもよい。 In recent years, various information input / output functions have been enhanced in convenience stores, etc., and services that meet the characteristic needs of each individual by identifying the characteristics of the purchase history and payment history of a specific individual, centered on these stores, are available. Has been built. Therefore, in cooperation with this information, the above-mentioned information provision may be reflected in the purchasing behavior. For example, for a person whose blood pressure is becoming high, purchasing advice or dietary advice to prevent hypertensive symptoms may be given at a store.
　このような特徴情報も時系列で整理し、図２に示す表において、情報判定機器２が出力するデータ（検査データ、生体データ、バイタルデータ、検体データなどを総じて「時系列データ」として表記）がどのような時間的変化、推移となったかと同様の考え方で情報に入れ込んで扱ってもよい。どのような生体データ変化で、それがどのような生活習慣（食事の傾向など）の患者（患者ＩＤとして説明）が、どのような医療機関等と関連したかがまとめられたデータ群となる。つまり、このようなデータベースを利用する判定を行うことによって、特定の傾向の患者に対して、かつて類似の傾向であった患者が回復した例を検索可能となり、どのような生活改善やどのような医療機関へ受診や服薬や治療が好ましいかを伝達できる情報伝達装置および情報伝達方法が提供可能となる。 Such feature information is also organized in chronological order, and in the table shown in FIG. 2, the data output by the information determination device 2 (test data, biological data, vital data, sample data, etc. are generally expressed as "time series data"). You may treat it by incorporating it into the information in the same way as what kind of temporal change or transition the data has changed. It is a data group that summarizes what kind of biological data changes and what kind of lifestyle (dietary tendency, etc.) patients (explained as patient ID) are related to what kind of medical institution. In other words, by making a judgment using such a database, it becomes possible to search for cases in which patients with a specific tendency have recovered, and what kind of life improvement and what kind of life improvement. It will be possible to provide information transmission devices and information transmission methods that can convey to medical institutions whether consultation, medication, or treatment is preferable.
　この情報伝達装置等は、皮膚科に行けばよいとか、消化器系医院に行けばよいといった、類型的なアドバイスを提供するのではなく、その人がアクセスしやすい範囲内で、類似の特徴の改善策を提供する。つまり、ここでのデータベース（ＤＢ）には、各個人の生体情報の変化パターンと、健康、医療関係行動の関係が時間軸で整理されて関連づけられている。また、医療機関や健康関連施設や周辺の店舗の特徴情報が特定のタイミングで更新され、刷新されると共に、生体情報変化時点で、どのような状況であったかが記録されていて検索可能である。 This information transmission device, etc. does not provide typographical advice such as going to a dermatologist or a digestive system clinic, but has similar characteristics within the range that the person can easily access. Provide remedial measures. That is, in the database (DB) here, the change pattern of the biological information of each individual and the relationship between health and medical-related behaviors are organized and associated with each other on the time axis. In addition, the characteristic information of medical institutions, health-related facilities, and stores in the vicinity is updated and renewed at a specific timing, and the situation at the time of the change in biometric information is recorded and can be searched.
　上述の特徴情報を提供するため、図２の中で、「その他」として示した項目としては、医療機関単独ではなく周辺の状況や、それらの過去の情報などを含めてもよい。つまり、特定の仕様の検査機器を用いて時系列的に検査データを取得し、該検査データの変化パターン情報を取得し、変化パターン情報に対し、改善効果のあった医療機関等における特徴情報（影響イベント情報）をアノテーションした結果を教師データとして学習させた推論モデルに対し、特定の人物の特定の仕様の検査機器を用いて時系列的に得た検査データを入力して推論した結果を伝達することを特徴とする情報伝達方法が提供可能となる。 In order to provide the above-mentioned feature information, the items shown as "others" in FIG. 2 may include not only the medical institution alone but also the surrounding situation and their past information. That is, test data is acquired in chronological order using a test device having specific specifications, change pattern information of the test data is acquired, and characteristic information (characteristic information) in a medical institution or the like that has an improvement effect on the change pattern information ( The inferred result is transmitted by inputting the inspection data obtained in time series using the inspection equipment of the specific specifications of a specific person to the inference model in which the result of annotating the impact event information) is trained as teacher data. It becomes possible to provide an information transmission method characterized by doing so.
　上述の情報伝達方法では、言い換えると、まず、特定の仕様の検査機器を用いて特定期間の時系列的に検査データを取得する。そして時系列的に取得した検査データから、検査データの変化パターン情報を取得する。この変化パターン情報について、特定期間内で変化パターンが変化した時点に対応する時点における影響イベント情報を検出する。影響イベント情報は、医療機関等において改善効果があったと認められた場合に、改善に寄与したイベントに関する情報である。影響イベント情報を検出すると、この検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する。推論モデルが生成されると、特定の人物の特定の仕様の検査機器を用いて、特定期間の幅に類似する期間の間に時系列的に得た検査データを、推論モデルに入力して影響イベントを推論する。推論結果を得ると、この推論結果をユーザ等に伝達する。 In the above-mentioned information transmission method, in other words, first, inspection data is acquired in chronological order for a specific period using an inspection device having a specific specification. Then, the change pattern information of the inspection data is acquired from the inspection data acquired in time series. Regarding this change pattern information, the influence event information at the time corresponding to the time when the change pattern changes within a specific period is detected. The impact event information is information on the event that contributed to the improvement when it is recognized that the improvement effect was achieved by the medical institution or the like. When the impact event information is detected, an inference model is acquired in which the result of annotating the detected impact event information is trained as teacher data. When the inference model is generated, the inference model is affected by inputting the inspection data obtained in time series during a period similar to the width of the specific period using the inspection equipment of the specific specification of a specific person. Infer the event. When the inference result is obtained, the inference result is transmitted to the user or the like.
　なお、推論モデルを生成するためには、適正規模の教師データからなる学習用母集団を作成しなければならない。このために、多数の個人の情報を収集して学習用母集団を作成する。この場合、検査機器の種類や、また検査を行った期間等の情報を含めて収集し、これらの情報毎に学習用母集団を作成するとよい。 In addition, in order to generate an inference model, it is necessary to create a learning population consisting of teacher data of an appropriate scale. For this purpose, a large number of individual information is collected to create a learning population. In this case, it is advisable to collect information such as the type of inspection equipment and the period during which the inspection was performed, and create a learning population for each of these pieces of information.
　また、ここでの時系列パターンや、この時系列パターンを記録するデータベースは、言い換えると、下記のような特徴を有するものであると言える。つまり、特定の仕様の検査機器を用いて時系列的に検査データを取得し、かつ、この時系列パターンの時間軸に合わせて、服薬や生活改善の開始等、特定の改善タイミング情報を有するデータベースである。つまり、特定の改善タイミング以降、該検査データの変化パターン情報も取得し、変化パターンが確かに改善されているという特徴が表されている例をも含み得るデータベースになっていればよい。改善タイミング以前で、将来の患者の変化パターンの類似性を判定可能なものであって、その患者候補の生体情報変化パターンが改善する可能性がある何らかの情報を含んでいれば、多くの人に利用されて利便性のあるデータベースとなりうる。 In other words, the time-series pattern here and the database that records this time-series pattern can be said to have the following characteristics. In other words, a database that acquires test data in chronological order using a test device with specific specifications and has specific improvement timing information such as taking medication and starting life improvement according to the time axis of this time series pattern. Is. That is, it suffices if the database can include an example in which the change pattern information of the inspection data is acquired after the specific improvement timing and the feature that the change pattern is surely improved is also included. Before the timing of improvement, if it is possible to determine the similarity of the change pattern of the patient in the future and it contains some information that the change pattern of the biological information of the patient candidate may be improved, many people will be asked. It can be used and become a convenient database.
　勿論、医療機関の個々の情報として、保有機器情報は有効であり、保有設備としては、体温計、聴診器、血圧計、体重計、体脂肪計のような手軽なものから、Ｘ線検査装置、内視鏡、超音波検査装置、ＣＴ、ＭＲＩといった据え置き型のもの、あるいは様々な伝染性疾患の検査装置や試薬等まで、ここでは想定している。なお、かつ、これらの装置が機種ごとに異なる性能や機能を有している場合は、それも含めて識別管理可能としている。内視鏡にも硬性鏡、軟性鏡の他、特殊光観察が出来るものと出来ないものもある。これによって、その施設が出来る検査の特徴や種別や限界などが判定される。 Of course, as individual information of medical institutions, possession device information is effective, and possession equipment includes simple items such as thermometers, stethoscopes, sphygmomanometers, weight scales, and body fat scales, as well as X-ray examination devices. Here, we assume endoscopic ultrasonography equipment, stationary equipment such as CT and MRI, and examination equipment and reagents for various infectious diseases. In addition, if these devices have different performances and functions for each model, they can be identified and managed including them. In addition to rigid and flexible endoscopes, some endoscopes can and cannot observe special light. As a result, the characteristics, types, limits, etc. of the inspections that the facility can perform are determined.
　処置具としても、様々な設備がありえ、それによって術式が異なる医療機関もあるので、これらの情報を整理し、管理することによって、判定可能としてもよい。検査装置があっても、必要な試薬や治療薬や治療装置などがない可能性もあるので、様々な検査試薬や治療機器、治療用薬剤、さらには包帯・ガーゼ、注射器、注射針、点滴などの医療・救急・衛生用品の在庫情報などの有無を管理できるようにしてもよい。その他、松葉づえや車いす等、患者の補助装置等に関する情報も管理できるようにしてもよい。また、バリアフリーの状態や、周辺宿泊施設や店舗、交通手段の情報も有益であり、公共交通機関が利用し易いかどうかや、入退院時に世話を行う家族に対する利便性などは、特定の疾病や年齢の患者には最も有益な事項となりうる。 As a treatment tool, there may be various facilities, and some medical institutions have different surgical procedures, so it may be possible to make a judgment by organizing and managing this information. Even if there is a test device, there is a possibility that the necessary reagents, therapeutic agents, therapeutic devices, etc. are not available, so various test reagents, therapeutic devices, therapeutic agents, bandages / gauze, syringes, needles, intravenous drip, etc. It may be possible to manage the presence or absence of inventory information of medical / emergency / sanitary supplies. In addition, information on assistive devices for patients such as crutches and wheelchairs may be managed. In addition, information on barrier-free conditions, surrounding accommodation facilities, stores, and transportation is also useful, and whether public transportation is easy to use and convenience for families who take care of them at the time of admission and discharge are related to specific illnesses. It can be the most beneficial for patients of age.
　保有設備の欄に、保有設備や用品、消耗品などの在庫等の情報を含めて施設の特徴を記録するようにしておけば、これらの情報をもとに有効なガイドを出すことが出来る。医師や検査の技師の性別や、個室の有無なども患者にはありがたい情報となるので、併せて表示、ガイド可能にデータベースで管理してもよい。病院等の施設にあるトイレが、バイタル情報や検体情報等、取得可能であるかや、 きちんと校正されて、信頼性の高いトイレか、あるいは検尿、検便に対応した施設であるか、 また、バイタル情報を取得可能な、患者の健康状態見守り用の貸し出し機器があるかどうか等の情報も施設情報として取得する方が好ましい。 診察に来院した患者のみならず、施設での校正されたトイレでの情報と、 自宅のトイレとの誤差、自宅トイレと検便との誤差情報を取得するために施設を訪れるユーザにも対応できる。 If you record the characteristics of the facility including information such as inventory of owned equipment, supplies, consumables, etc. in the column of owned equipment, you can issue an effective guide based on this information. Since the gender of doctors and examination technicians and the existence of private rooms are also useful information for patients, they may be displayed and guided in a database. Whether the toilets in hospitals and other facilities can obtain vital information, sample information, etc., whether they are properly calibrated and highly reliable toilets, or whether they are facilities that support urinalysis and stool tests, and whether they are vitals. It is preferable to acquire information such as whether or not there is a rental device for monitoring the patient's health condition, which can acquire information, as facility information. Not only patients who come to the clinic, but also users who visit the facility to obtain information on the calibrated toilet at the facility, the error between the toilet at home, and the error between the toilet at home and the stool test can be handled.
　ここでは、病院、クリニック等において診療を受ける例で説明しており、このような健康状態に影響がある行動（身体によい食べ物やサプリメントの摂取や、市販薬の服用などもここに含むことを想定している）や環境変化（転地療法や室内の環境改善などで健康状態が変わりうる）を健康対応イベントと呼び、特定の健康対応イベントが健康（ここでは時系列生体データパターン）にどう影響するかが判定できるシステムを構築しておく。このシステムは、人や医師が意識していなかったような健康を改善するもの、害するものが何かを判定可能にできる。つまり、特定の健康対応イベントの入力（これは手動でも自動でもよい）によって、そのイベントに先立って得ていた（しばらく効果がない場合もあるので、厳密に先立つ必要はない）時系列的な検査データの変化パターン情報が、結局、どうなるかを判定できる。イベントには、前述のように、単なるイベント名やタイミング情報の他、この健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報（構成要因）があると、意識されていなかった重要要因をあぶりだすことが可能となる。単に、皮膚科受診ではなく、そこでどのような検査を受けたかなどが重要情報となることもある。その検査が受けられるのは皮膚科以外の、例えば小児科でも可能かもしれない。 Here, the example of receiving medical treatment at a hospital, clinic, etc. is explained, and such behaviors that affect the health condition (intake of healthy foods and supplements, taking over-the-counter medicines, etc. are also included here. (Assumed) and environmental changes (health status can change due to relocation therapy or indoor environment improvement) are called health response events, and how specific health response events affect health (here, time-series biological data patterns). Build a system that can determine whether to do it. This system can determine what improves or harms health that people and doctors were not aware of. In other words, by entering a particular health response event (which can be manual or automatic), a time-series test that was obtained prior to that event (it may not be effective for some time, so it does not need to be strictly preceded). It is possible to determine what will happen to the data change pattern information in the end. As mentioned above, it is conscious that the event has not only the event name and timing information but also the equipment and / or equipment and / or environmental information (constituent factors) at the time of this health response event. It will be possible to reveal important factors that have not been done. Sometimes important information is not just a dermatologist's visit, but what kind of test was done there. The test may be available in non-dermatologists, such as pediatrics.
　このような健康対応イベントが何であったが分かれば、健康管理に役立つと思われるが、現実には種々の健康対応イベントがあり、いずれの健康対応イベントが重要要因であったかを知ることが困難である。そこで、本実施形態においては、以下のような推論モデルを生成し、推論を行うようにしている。まず、健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報をアノテーションした結果を教師データとし、この教師データを用いて学習させることによって推論モデルを取得する。推論モデルを取得すると、前述したように、特定の人物の特定の仕様の検査機器を用いて時系列的に得た検査データを、推論モデルに入力して設備、および／または備品、および／または環境の情報を推論すれば、先の重要要因を得ることが出来る。この推論結果（重要要因）から特定の人物用にカスタマイズした健康対応イベントを、特定の人物に伝達すれば、わざわざ遠くの病院まで行かなくとも、近所もクリニックで済む場合や、生活の改善などで済む場合がありえる。 It would be useful to know what such a health response event was, but in reality there are various health response events, and it is difficult to know which health response event was an important factor. is there. Therefore, in the present embodiment, the following inference model is generated to perform inference. First, the result of annotating the equipment and / or equipment and / or environment information at the time of the health response event is used as teacher data, and the inference model is acquired by training using this teacher data. When the inference model is acquired, as described above, the inspection data obtained in time series using the inspection equipment of the specific specifications of the specific person is input to the inference model, and the equipment and / or equipment and / or By inferring environmental information, the above important factors can be obtained. If you convey a health response event customized for a specific person from this inference result (important factor) to a specific person, you can go to a clinic in the neighborhood without having to go to a distant hospital, or improve your life. It may be done.
　このように、上述の情報伝達システムによれば、健康対応イベントの要素まで含めた推論が出来る。なお、上述のシステムでは、本当にその人の健康状態が改善したかどうかまでは情報を付与していないが、多くの人の取る行動などは推論できる。また、健康状態が改善した場合のみ、特定の健康対応イベントの入力（これは手動でも自動でもよい）を行ってもよい。改善は検査データ変化などでも判定できるし、またアンケートに基づいて判定しても良い。つまり、病院が診療後に、患者にアンケートメールを送信し、気になる症状が改善したという答えが返って来た時のみ、健康対応イベント入力が行われるようなシステムでもよい。 In this way, according to the above-mentioned information transmission system, it is possible to make inferences including the elements of health response events. In the above system, information is not given as to whether or not the person's health condition has really improved, but the actions taken by many people can be inferred. In addition, a specific health response event (which may be manual or automatic) may be input only when the health condition is improved. Improvement can be judged by changes in inspection data, etc., or may be judged based on a questionnaire. In other words, the system may be such that the hospital sends a questionnaire mail to the patient after the medical treatment, and the health response event is input only when the answer that the symptom of concern has improved is returned.
　健康状態が改善したか否かは検査データ変化などでも判定できると書いた。具体的には、検査機器を用いて得た生体情報の変化パターン情報に対し、特定の改善タイミング情報を変化パターン情報の時間情報と同様の時間情報として含むことによって、改善タイミング情報を参考に変化パターンの部分を比較可能にした情報を有するデータベースを作成しておけば、それが可能となる。このデータベースを用いて、特定の人物（アドバイス情報が欲しい人）の変化パターン部分を他の人の改善タイミング情報を参考に変化パターン部分と比較して、改善タイミング情報における改善情報を判定し、判定された改善情報を特定の人物に伝達すればよい。また、ここでは、改善した場合を強調して説明したが、こうしたイベントで体調が悪くなった、というような推論の用途にも応用できることは言うまでもない。 I wrote that it can be judged by changes in test data whether or not the health condition has improved. Specifically, by including specific improvement timing information as time information similar to the time information of the change pattern information with respect to the change pattern information of the biological information obtained by using the inspection device, the change is made with reference to the improvement timing information. This is possible if you create a database that contains information that makes the pattern parts comparable. Using this database, the change pattern part of a specific person (person who wants advice information) is compared with the change pattern part with reference to the improvement timing information of other people, and the improvement information in the improvement timing information is judged and judged. The improvement information made may be transmitted to a specific person. In addition, although the explanation emphasizes the case of improvement here, it goes without saying that it can also be applied to inference applications such as when one feels sick due to such an event.
　つまり、特定の仕様の（生体データ）検査機器を用いて特定の期間にわたって時系列的に複数の検査データを取得して得たデータの変化パターン情報を有し、この特定の期間内におけるデータ取得元に関する複数のイベント情報を、それぞれ何時起こったかのタイミングまで関連付けて記録可能なデータベースがあれば、トレンド変化に影響したイベントが抽出できるはずである。そこで、この影響があったイベントの情報提供を可能とする回路やプログラムやシステムとすることで、他の人に有益な情報を提供しうる。健康状態に影響を与えたイベントのあった環境や状況やその他の構成要素情報に基づいて、イベント情報を細分化すれば、情報提供する人の立場に置き換えたカスタマイズ情報として提供することができる。例えば、寒い日に人混みに行った、といった要素に分解すれば、東京駅に行ったという情報より一般化でき、大阪に住む人で、体調が悪そうな人には、今日は大阪駅には行くな、といった情報に変換して伝えられる。また、検査データ変化パターンと、影響イベントのアノテーションによって得られた多数の人のケースの教師データを集め、学習し推論モデルを作成してもよい。この場合、入力を変化パターン、出力を影響イベント要素とすればよい。 That is, it has change pattern information of data obtained by acquiring a plurality of inspection data in a time series using a (biological data) inspection device having a specific specification, and data acquisition within this specific period. If there is a database that can record multiple event information about the original in association with the timing of when it happened, it should be possible to extract the events that affected the trend change. Therefore, it is possible to provide useful information to other people by providing circuits, programs, and systems that can provide information on events that have been affected by this. If the event information is subdivided based on the environment and situation of the event that affected the health condition and other component information, it can be provided as customized information replaced by the position of the person who provides the information. For example, if you break it down into factors such as going to a crowd on a cold day, it can be generalized from the information that you went to Tokyo Station. It is converted into information such as "Don't go" and transmitted. In addition, the test data change pattern and the teacher data of a large number of cases obtained by annotating the influence event may be collected and learned to create an inference model. In this case, the input may be the change pattern and the output may be the influence event element.
　また、ＤＢ部８のデータベースには、症例数Ｎｃ１～Ｎｃ６等を記録しておく。関連疾病ｌｘについて、その医院が処置した症例数があれば、ユーザが医療施設を選択する際の参考になり、また推論等によって推奨医療施設を表示する際にも使用することができる。また、ＤＢ部８のデータベースのその他の欄には、予後の情報、入院期間の情報、隣接薬局情報（例えば、品揃え情報を含んでいてもよい）等の情報を記録できるようにしてもよい。 In addition, the number of cases Nc1 to Nc6 and the like are recorded in the database of the DB unit 8. If there is a number of cases treated by the clinic for the related disease lp, it can be used as a reference when the user selects a medical facility, and can also be used when displaying a recommended medical facility by inference or the like. In addition, information such as prognosis information, hospitalization period information, and adjacent pharmacy information (for example, assortment information may be included) may be recorded in other columns of the database of the DB unit 8. ..
　制御部１の検索部１ｆは、図２に示すようなデータベースにアクセスし、必要な情報を取りだす機能を有する。図２において、患者ＩＤは、ＩＤ判定部１ｂにおいて、個々の患者（ユーザ）に付与されている識別用の符号で、患者を特定する。時系列データは、情報判定機器２によって測定された個々のデータである。例えば、データＤｙ１（ｔ１１）は、日時ｔ１１に測定された種別ｙ１のデータである。データ種別ｙは、例えば、体温、血圧等のバイタル情報や、大便等の検体情報である。来院日時Ｔｈは、患者が医療機関で受診したときの日時情報を示す。関連疾病ｌｘは、患者ＩＤの時系列情報から想定される疾病名に関する情報を示す。 The search unit 1f of the control unit 1 has a function of accessing the database as shown in FIG. 2 and extracting necessary information. In FIG. 2, the patient ID is an identification code given to each patient (user) in the ID determination unit 1b to identify the patient. The time series data is individual data measured by the information determination device 2. For example, the data Dy1 (t11) is the data of the type y1 measured at the date and time t11. The data type y is, for example, vital information such as body temperature and blood pressure, and sample information such as stool. The date and time of visit Th indicates information on the date and time when the patient visited a medical institution. The related disease lp indicates information on the disease name assumed from the time series information of the patient ID.
　医院Ｈ１～Ｈ４は、医療施設を示し、患者ＩＤによって特定される者が受診した医療施設を示す。診療科Ｄｐ１～Ｄｐ３は、患者ＩＤによって特定される者が受診した医療施設における診療科名を示す。保有設備Ｍｏｄは、患者ＩＤによって特定される者に関連する保有設備である。症例数Ｎｃ１～Ｎｃ６は、各医院・各診療科において、扱った症例の数である。ユーザに推奨設備を表示する際に、症例数を考慮してもよく、また、ユーザが医院等を選択する際の参考情報とするために表示するようにしてもよい。 Clinics H1 to H4 indicate medical facilities, and indicate medical facilities where a person specified by a patient ID has visited. The clinical departments Dp1 to Dp3 indicate the names of clinical departments in the medical facility where the person specified by the patient ID visited. The possession equipment Mod is the possession equipment associated with the person identified by the patient ID. The number of cases Nc1 to Nc6 is the number of cases handled in each clinic / clinical department. When displaying the recommended equipment to the user, the number of cases may be taken into consideration, or the recommended equipment may be displayed for reference information when the user selects a clinic or the like.
　なお、図２においては、施設別の保有機器一覧の記録部８ａと、施設別のＩＤと来院情報一覧の記録部８ｂを混合した一覧表で示している。図２において、医院Ｈ１～Ｈ４毎に、保有設備Ｍｏｄ１～Ｍｏｄ５の一覧表を作成すれば、記録部８ａの記録内容となる。また、患者ＩＤ毎に、来院日時等の一覧表を作成すれば、記録部８ｂの記録内容となる。また、患者ＩＤ毎に、時系列データのみを記録しておけば、治療を受けた医療施設に関わらず、個々の患者毎の履歴データを作成することができる。この履歴データは、後述する図４のＳ９、図７のＳ６における推論を行う際に使用することも可能である。 Note that FIG. 2 shows a list in which the recording unit 8a of the list of owned devices by facility and the recording unit 8b of the ID and visit information list by facility are mixed. In FIG. 2, if a list of owned equipment Mod1 to Mod5 is created for each of the clinics H1 to H4, the recorded contents of the recording unit 8a will be obtained. Further, if a list of the date and time of visit is created for each patient ID, the recorded contents of the recording unit 8b can be obtained. Further, if only time-series data is recorded for each patient ID, historical data for each individual patient can be created regardless of the medical facility that received the treatment. This historical data can also be used when making inferences in S9 of FIG. 4 and S6 of FIG. 7, which will be described later.
　次に、図３を用いて、情報判定機器２等によって時系列的に取得される生体情報（検査データ）について説明する。図３は、検査データを用いて作成したグラフである。前述したように、ＤＢ部８に記録されている時系列的な検査データの例を図２に示している。図２には、患者ＩＤごとに、時系列で整理された検査データが記録されており、図３は図２に示す検査データをグラフに示したものである。図３において、横軸は時間Ｔを示し、縦軸は図２における時系列データをプロットして示したものである。これは前述のように、検査データ、生体データ、バイタルデータ、検体データであって、これらのいずれかを検査する機器の検査出力結果を数値Ｄに表したものである。例えば、大便の赤色の程度を示す値である。 Next, with reference to FIG. 3, biological information (examination data) acquired in time series by the information determination device 2 or the like will be described. FIG. 3 is a graph created using inspection data. As described above, FIG. 2 shows an example of time-series inspection data recorded in the DB unit 8. In FIG. 2, examination data organized in chronological order is recorded for each patient ID, and FIG. 3 is a graph showing the examination data shown in FIG. In FIG. 3, the horizontal axis shows the time T, and the vertical axis shows the time series data in FIG. 2 plotted. As described above, this is test data, biological data, vital data, and sample data, and the test output result of the device that inspects any of these is represented by a numerical value D. For example, it is a value indicating the degree of red color of stool.
また、図３では、来院日時等もシステム的に自動で更新されることを想定している。来院日時は、当然、複数あり得るが、煩雑さを避けるために、単純化して、例えば、特定診療科の初診の日時でもよい。図３（ａ）に示す例は、後述するように、時系列データが健康悪化の方向に向かって変化し、やがてユーザが通院に至ったケースであるので、図２で説明した各患者のパターンに最も近い。図３に示すような状況の患者には、図２で説明したような情報も利用して、あとどれ位で、どのような診療科のある病院に行くことになるかを推論した結果を提供することが可能となっている。なお、この例以外にも、図３（ｂ）のように、すでに他の兆候を自覚しているために通院しており、バイタルデータが得られる場合でも、図２に示すような表が作成できる。ただし、まったく通院していなくてもバイタルデータだけがある人もいるので、この場合は、図２の表には当てはまらない。 Further, in FIG. 3, it is assumed that the date and time of visit and the like are automatically updated systematically. Of course, there may be a plurality of visit dates and times, but in order to avoid complication, for example, the date and time of the first visit of a specific clinical department may be used. As will be described later, the example shown in FIG. 3A is a case in which the time-series data changes in the direction of deterioration of health and the user eventually goes to the hospital. Therefore, the pattern of each patient described in FIG. 2 Closest to. For patients in the situation shown in FIG. 3, the information as explained in FIG. 2 is also used to provide the result of inferring how long and what kind of clinical department the patient will go to. It is possible to do. In addition to this example, as shown in Fig. 3 (b), even if the patient is already aware of other signs and is going to the hospital and vital data can be obtained, a table as shown in Fig. 2 is created. it can. However, this does not apply to the table in FIG. 2 because some people have only vital data even if they do not go to the hospital at all.
　先に説明したように、図３（ａ）は、これから通院するであろうと推測されるケースである。図３（ａ）に示すグラフは、現在、通院していないユーザの検査データ（機器データ）の時系列的に変化を示す。この時系列的な検査データから、特定の検査結果（特定情報）を取得した時に、通常、医療機関に来院するかの情報が得うることができる。そこで、時系列的な検査データに基づいて、病院に行くほど悪化する前に、ユーザが受け取ることによって、自身の健康状態を把握できるような健康情報をガイドすることが可能となる。例えば、図３（ａ）においては、時刻Ｔ１における検査データの場合に、時間＋ΔＴが経過した時刻Ｔｃに、医療機関を訪れることを推論することができる。すなわち、ＤＢ部８に、検査データ、医療機関情報（医院名、診療科、日時情報）等が蓄積されていれば、医療機関で診療を受けるまでの期間が推測できる。 As explained earlier, Fig. 3 (a) is a case in which it is presumed that the patient will go to the hospital in the future. The graph shown in FIG. 3A shows changes in time series of examination data (equipment data) of users who are not currently visiting the hospital. From this time-series test data, it is possible to obtain information on whether or not to visit a medical institution when a specific test result (specific information) is obtained. Therefore, based on the time-series test data, it is possible to guide the health information so that the user can grasp his / her health condition by receiving it before it gets worse as he / she goes to the hospital. For example, in FIG. 3A, in the case of the test data at time T1, it can be inferred that the medical institution is visited at time Tc when time + ΔT has elapsed. That is, if the examination data, medical institution information (clinical name, clinical department, date and time information) and the like are accumulated in the DB unit 8, the period until receiving medical treatment at the medical institution can be estimated.
　図３（ｂ）は、既に通院している場合であり、治療以外の要因で、通院中に悪化したケースである。図３（ｂ）に示すグラフは、病気で通院している人が、時刻Ｔｃ１、Ｔｃ２において、特定情報が出現すると、医院で治療を受ける例である。図３（ｂ）に示すような時系列的な検査データは、このような状況を学習するに、十分利用が可能である。この例は、「この数値の人は普通、自分では治療できない」という趣旨のガイドに有効である。さらなる悪化を未然に防止できる情報となって有効である。 Fig. 3 (b) shows a case in which the patient has already visited the hospital and has deteriorated during the hospital visit due to factors other than treatment. The graph shown in FIG. 3B is an example in which a person who goes to the hospital due to illness receives treatment at a clinic when specific information appears at times Tc1 and Tc2. The time-series inspection data as shown in FIG. 3B can be sufficiently utilized to learn such a situation. This example is useful for a guide to the effect that "people with this number usually cannot be treated on their own." It is effective as information that can prevent further deterioration.
　図３（ｃ）は、医院に行く必要なないケースである。このケースでは、検査データＤは、所定値（グラフ中破線で示す）よりも低く、医院に行く必要がない。この場合、図２に示したＤＢ部８のデータベースにおいて、来院日時の欄は空欄となる。 Figure 3 (c) shows a case where it is not necessary to go to the clinic. In this case, the test data D is lower than the predetermined value (indicated by the broken line in the graph) and there is no need to go to the clinic. In this case, in the database of the DB unit 8 shown in FIG. 2, the column of the visit date and time is blank.
　図２に示すデータベース（ＤＢ部８に記録されている）において、その来院した医院や診療科の情報、保有設備の情報などが整理されて記録されている。このため、設備の事まで考えが及ばない患者にも、最適な施設を推奨することが可能となる。このデータベースは、取得情報の種別（便器の潜血検査情報）と医院と保有設備Ｍｏｄの関係を保持すればよく、患者別時系列データは別のデータベース管理でも良い。また、複数のＤＢを検索し、検索結果を整理することによって、ＤＢ部８に記録されるデータベースに相当する情報が得られる構成にしてもよい。 In the database shown in FIG. 2 (recorded in DB section 8), information on the clinics and clinical departments visited, information on owned equipment, etc. are organized and recorded. For this reason, it is possible to recommend the optimal facility even for patients who do not even think about the facility. This database may hold the relationship between the type of acquired information (toilet bowl occult blood test information), the clinic, and the owned equipment Mod, and the patient-specific time-series data may be managed in a separate database. Further, by searching a plurality of DBs and organizing the search results, information corresponding to the database recorded in the DB unit 8 may be obtained.
　図３は、ＤＢ部８に記録された患者ごとの時系列情報をグラフによって示しており、横軸が時間で縦軸が取得情報を数値化したものである。このため、２次元にビジュアルな情報となっている。２次元の図になっていることから、次の二つのことが言える。まず図であることから、画像判定と同様に扱うことができ、画像認識の推論モデルのような汎用的で構築しやすいＡＩチップもしくはシステムを簡単に流用でき、推論を容易に実現することができる。また、横軸が時間であることから、身体的な情報の時間変化の情報を有効利用でき、予測などを簡単にできる。また、繰り返し、類似のデータを取得しているので、データ数が集まりやすく、個々のデータに多少の誤差があっても、傾向としては他の人のデータ群とパターンを比較しての判定がしやすい。 FIG. 3 is a graph showing the time-series information for each patient recorded in the DB unit 8, where the horizontal axis is time and the vertical axis is the numerical value of the acquired information. Therefore, the information is two-dimensionally visual. Since it is a two-dimensional diagram, the following two things can be said. First, since it is a diagram, it can be handled in the same way as image judgment, and a general-purpose and easy-to-build AI chip or system such as an inference model for image recognition can be easily diverted, and inference can be easily realized. .. In addition, since the horizontal axis is time, it is possible to effectively use the time-varying information of physical information, and it is possible to easily make a prediction. In addition, since similar data is acquired repeatedly, the number of data is easy to collect, and even if there are some errors in individual data, the tendency is to compare the pattern with other people's data groups to make a judgment. It's easy to do.
　また、この時間軸上には、後述の日々の日常の生活上のイベント等による生活リズムの情報を反映させる事が出来る他、その人の健康に影響したり反映されたりする健康、医療上のイベントもまた反映させることが出来るようにするのが良い。例えば、何か運動を始めた時期等はこの時間軸の上におけるイベントであり、また何か健康食品の摂取を始めることによって、体調がよくなった場合等も、この時間軸に関連付けて、検査データの変化と関連付けることも可能である。図３（ａ）、（ｂ）、（ｃ）では単純化して、通院があったかどうかという例でわかりやすく図示したが、通院した医療機関の情報等までを情報として入力可能であれば、先の検査データ列のパターン変化と、これらのイベントとの関係の影響までも含めて統計的に判定することが可能となる。 In addition, on this time axis, it is possible to reflect information on the daily rhythm of daily life events, etc., which will be described later, as well as health and medical aspects that affect or reflect the health of the person. It is good to be able to reflect the event as well. For example, the time when you started exercising is an event on this time axis, and even if you feel better by starting to take some health food, you can check it in relation to this time axis. It can also be associated with changes in data. In FIGS. 3 (a), (b), and (c), it is simplified and illustrated in an easy-to-understand manner with an example of whether or not there was a hospital visit. It is possible to statistically judge the pattern change of the inspection data string and the influence of the relationship with these events.
　機械学習等を行う場合には、例えば、検査データが良い方向に変化したもの（例えば、対応する人の特定期間の時系列データ群、時系列検査情報であり、さらに健康関係の情報を、検査データとは識別可能に加えたものでもよい）に対し、「良いデータ変化」というアノテーションを行ってもよい。 When performing machine learning, for example, inspection data is changed in a positive direction (for example, time-series data group of a corresponding person for a specific period, time-series inspection information, and health-related information is inspected. The data may be added so that it can be identified), and the annotation "good data change" may be performed.
　一方、対象者（ユーザ）が最終的に医療機関を頼らなければならなかった場合は、その診療科や保有機器などを検査データ等にアノテーションする。これらを教師データとして学習させ、この場合、さらに「通院時、自覚症状」なども合わせて教師データ化してもよい。この工夫を行うと、測定した時系列データに無関係の病気で行った病院などについては排除した推論が可能となる。例えば、血圧などの変化で慢性疾患などを予測するサービスに本実施形態を応用した場合、たまたま、花粉症の時期に耳鼻科に行く人が増えた場合などの情報が混入して、耳鼻科が紹介されてしまうような事を防ぐことが好ましい。人がアノテーションする時には、このような誤情報が混入する事は起こりにくい。しかし、アノテーションを一部自動化する場合などは、「通院時・自覚症状」が「くしゃみ」で、行った医療機関が「耳鼻科」である場合には、この医療機関は血圧から導かれる医療機関には含めないようにする、といったフィルタリングを行うとよい。 On the other hand, if the target person (user) finally has to rely on a medical institution, annotate the clinical department, possessed equipment, etc. in the examination data. These may be learned as teacher data, and in this case, "at the time of hospital visit, subjective symptoms" and the like may also be converted into teacher data. By taking this measure, it is possible to make inferences that exclude hospitals that have been treated for illnesses unrelated to the measured time-series data. For example, when this embodiment is applied to a service that predicts chronic diseases due to changes in blood pressure, etc., information such as when the number of people who happen to go to otolaryngology during the period of pollinosis increases, and otolaryngology It is preferable to prevent such things as being introduced. When a person annotates, it is unlikely that such false information will be mixed in. However, when partially automating the annotation, if the "at the time of hospital visit / subjective symptom" is "sneezing" and the medical institution that went to is "otolaryngology", this medical institution is derived from blood pressure. It is good to perform filtering such as not including it in.
　このように、多数のプロフィール情報と、プロフィール毎の検査データを取得する検査データ取得部と、プロフィール毎に診療された医療機関を記憶する記憶部と、検査データと、プロフィール情報と、診療医療機関情報を教師データとして学習された推論モデルによって、特定の対象者の検査データを入力として、診療医療機関情報を推論して対象者に伝達する伝達情報を決定する伝達情報決定部と、を有することを特徴とする情報伝達装置が提供可能となる。 In this way, a large amount of profile information, an examination data acquisition unit that acquires examination data for each profile, a storage unit that stores the medical institution that has been treated for each profile, examination data, profile information, and a medical institution for medical treatment. It has a transmission information determination unit that infers medical institution information and determines the transmission information to be transmitted to the target person by inputting the test data of a specific target person by the reasoning model learned from the information as teacher data. It becomes possible to provide an information transmission device characterized by.
　この時、教師データに使用する診療医療機関情報に、その機関の保有機器、保有設備や診療実績などの特徴を含めておけば、より詳しい情報が提供できる。特定の設備がある医療機関が推論される場合は、その特定の人にもその情報をシェアしてもよいし、その保有設備情報で、その人の住所の近隣で、その保有機器を有する医療機関を検索して情報としてもよい。 At this time, more detailed information can be provided if the medical institution information used for the teacher data includes features such as the equipment, equipment, and medical records owned by the institution. If a medical institution with a specific facility is inferred, the information may be shared with that specific person, or the medical facility possessing the facility can be used to provide medical care with the device in the vicinity of the person's address. You may search for an institution and use it as information.
　また、時系列情報には、揺らぎや頻度といった、生体特有の時間変化の特徴に関する情報を盛り込める。例えば、ユーザが、寝ている時、起きている時、朝と昼と夜、食前、食後、入浴前後といった情報を考慮しやすいという点があげられる。また、心拍や呼吸などは揺らぎが適当な方がリラックスしていて健康的という研究もある。必要に応じて、上述のようなタイミングと思われる時のデータのみを抜き出したり、特定の状況のものを省いたりして、これらのデータのみを利用する等、様々な使い方が可能である。 In addition, the time series information can include information on the characteristics of time changes peculiar to the living body, such as fluctuations and frequency. For example, it is easy for the user to consider information such as when sleeping, when waking up, morning and day and night, before meals, after meals, and before and after bathing. In addition, there is a study that it is more relaxed and healthier to have appropriate fluctuations in heart rate and breathing. If necessary, it can be used in various ways, such as extracting only the data at the time when it seems to be the timing as described above, omitting the data in a specific situation, and using only these data.
　また、図３において、履歴データを取得している期間が特定期間に相当し、この特定期間の間の検査データが抽出される。この抽出された時系列的な検査データが、推論エンジン７に入力され、推論エンジン７は推論によってアドバイスを出力し、対象者にアドバイスが提供される。また、特定の時間幅は、その時間幅の終了時点から後の未来に相当する時点で、何らかのアドバイスが出来るのに相応しいものであればよく、このアドバイス時点に対して遡った複数の情報が取得できる幅であれば良い。また、特定の時間幅は、規格化してもよいし、また必ずしも厳密にしたものでなくともよく、十分なデータ量が得られれば良い場合もある。各データの時間間隔なども重要な情報となるので、規則的な時間幅で得られた、離散的ではないものであった方が良い。しかし、時間幅内においてデータの測定時点が離散的であっても、データを補間で補って意味のあるデータが得られる程度の時間幅であるならば有効である。何らかの健康、医療関係情報に従って決められるものでもよい。 Further, in FIG. 3, the period during which the history data is acquired corresponds to the specific period, and the inspection data during this specific period is extracted. The extracted time-series inspection data is input to the inference engine 7, the inference engine 7 outputs advice by inference, and the advice is provided to the subject. In addition, the specific time width may be suitable for giving some advice at a time corresponding to the future after the end of the time width, and a plurality of information retroactive to the time of this advice is acquired. It should be as wide as possible. Further, the specific time width may be standardized, or may not necessarily be strict, and it may be sufficient if a sufficient amount of data can be obtained. Since the time interval of each data is also important information, it is better that the data is obtained with a regular time width and is not discrete. However, even if the measurement time points of the data are discrete within the time width, it is effective as long as the time width is such that the data can be supplemented by interpolation to obtain meaningful data. It may be determined according to some health or medical information.
　また、横軸の幅を適当に規定することによって、予測などの精度を推論モデルによって切り替えることも可能である。時間幅が１年程度であれば、数か月オーダーの予測が可能になり、時間幅が１週間程度では、数日オーダーの予測に向いており、病気の特徴に応じて適切な幅を変えることが出来る。例えば、腫瘍のように徐々に進行するものと、インフルエンザなど感染症のように急激に治るものと悪化するものを見極めなければならない場合では、適当な時間幅が異なる。つまり、本実施形態に係る情報伝達装置は、予め定められた時間幅で対象者の検査データの変化パターンを抽出し、時間情報と共に学習された推論モデルに従って、対象者への伝達情報を決定する伝達情報決定部を有する。 Also, by appropriately defining the width of the horizontal axis, it is possible to switch the accuracy of prediction etc. by the inference model. If the time width is about one year, it is possible to predict orders for several months, and if the time width is about one week, it is suitable for forecasting orders for several days, and the appropriate width is changed according to the characteristics of the disease. Can be done. For example, when it is necessary to distinguish between a tumor that progresses gradually and an infectious disease such as influenza that cures rapidly and worsens, the appropriate time range differs. That is, the information transmission device according to the present embodiment extracts the change pattern of the inspection data of the subject in a predetermined time width, and determines the transmission information to the subject according to the inference model learned together with the time information. It has a transmission information determination unit.
　次に、図４に示すフローチャートを用いて、情報伝達システムにおける検査結果の送信の動作について説明する。このフローは、主として、制御部１内のＣＰＵがメモリに記憶されたプログラムに従って、情報伝達システム全体を制御することによって、実行される。また、図４に示すフローは、図１に示したＤＢ部８における検索と、推論エンジン７による推論等の機能を個別に利用する場合を示す。どちらかの一方の機能を使用する場合や、両方を重ねて使用する場合もあり得るが、ここでは、最も簡潔な例を示している。 Next, the operation of transmitting the inspection result in the information transmission system will be described using the flowchart shown in FIG. This flow is mainly executed by the CPU in the control unit 1 controlling the entire information transmission system according to the program stored in the memory. Further, the flow shown in FIG. 4 shows a case where the search in the DB unit 8 shown in FIG. 1 and the inference by the inference engine 7 are individually used. You may want to use one of the functions, or you may use both of them in layers, but here is the simplest example.
　図４のフローを説明するにあたって、情報判定機器２として、便器に画像センサや顕微鏡のような拡大画像判定器、特殊な光の反射などを検出するセンサ、結晶性ナノワイヤーのアレイや、分子膜などの電気特性の変化を応用した嗅覚センサ、ガス成分センサ等が配置され、ユーザの排泄物の特徴を確認できる場合を想定して説明する。 In explaining the flow of FIG. 4, as the information judgment device 2, an image sensor, a magnifying image judgment device such as a microscope, a sensor for detecting special light reflection, an array of crystalline nanowires, and a molecular film are used as the information judgment device 2. The explanation will be made on the assumption that an olfactory sensor, a gas component sensor, etc. that apply changes in electrical characteristics such as the above are arranged and the characteristics of the user's excrement can be confirmed.
　図４に示すフローにおいて、表示する推奨施設は、ＤＢ部８に記録されたデータベースの中から検索する（Ｓ５Ｙｅｓ→Ｓ７参照）。このために、ＤＢ部８には、特定情報に基づいてユーザの検査や治療を行うことができる施設に関する情報のデータベースを構築しておく。しかし、データベースによっては、特定情報に基づいて施設に関する情報を検索することができない場合もある。その場合には、ユーザの履歴データを用いて推論を行うようにしている（Ｓ５Ｎｏ→Ｓ９参照）。 In the flow shown in FIG. 4, the recommended facility to be displayed is searched from the database recorded in the DB unit 8 (see S5Yes → S7). For this purpose, the DB unit 8 is constructed with a database of information on facilities that can perform user examinations and treatments based on specific information. However, depending on the database, it may not be possible to search for information about the facility based on specific information. In that case, the inference is performed using the user's history data (see S5No → S9).
　図４に示す検査結果送信のフローが開始すると、まず、ＩＤ毎に、センサ出力結果に基づいて判定する（Ｓ１）。ここでは、制御部１が通信制御部１ａを通じて情報判定機器２の出力を取得する場合や、情報取得機器２が送信したデータを制御部１が通信制御部１ａで受けとる場合がある。また、情報判定機器２が記録していたデータを特定のタイミングで通信制御部１ａを通じて制御部１が収集するような方法などを想定している。このとき、センサ出力結果に添付されたＩＤ毎に、センサ出力に基づいて、検査結果の判定を行う。センサとしては、色センサ、形状センサ、硬度センサ、嗅覚センサ、ガス成分センサ、特定の試薬添加時の色変化検出センサであり、イメージセンサの出力に基づいて、拡大観察画像による形状判定を行ってもよい。 When the flow of inspection result transmission shown in FIG. 4 starts, first, each ID is determined based on the sensor output result (S1). Here, the control unit 1 may acquire the output of the information determination device 2 through the communication control unit 1a, or the control unit 1 may receive the data transmitted by the information acquisition device 2 in the communication control unit 1a. Further, it is assumed that the data recorded by the information determination device 2 is collected by the control unit 1 through the communication control unit 1a at a specific timing. At this time, the inspection result is determined based on the sensor output for each ID attached to the sensor output result. The sensors are a color sensor, a shape sensor, a hardness sensor, an olfactory sensor, a gas component sensor, and a color change detection sensor when a specific reagent is added. Based on the output of the image sensor, the shape is determined by a magnified observation image. May be good.
　前述のユーザの排泄物の特徴を確認する場合には、例えば、潜血のある便などは色センサで判定が可能である。また、排せつの量や形状、硬さなどは、イメージセンサ・色センサによって判定してもよいし、特殊な染色を行って色の分布などを測定する方法でも良い。あるいは、対象物を拡大した画像で組成を検出してもよく、特定の時間、培養した結果を判定してもよい。例えば、便に混ざる血液が増えると赤血球の赤色が目立ってくるが、これを数値化すると、健康な場合との差異が分かる。ステップＳ１において、これらを検出する。 When confirming the characteristics of the user's excrement described above, for example, stool with occult blood can be determined by a color sensor. Further, the amount, shape, hardness, etc. of excretion may be determined by an image sensor / color sensor, or a method of measuring the color distribution by performing special dyeing may be used. Alternatively, the composition may be detected in a magnified image of the object, or the result of culturing for a specific time may be determined. For example, when the amount of blood mixed in the stool increases, the red color of red blood cells becomes conspicuous, but if this is quantified, the difference from the healthy case can be seen. These are detected in step S1.
　ステップＳ１において、制御部１が特定のプログラム等を用いた判断等によって、センサ出力結果の判定を行うと、次に、特定情報を得ることが出来たか否かを判定する（Ｓ３）。また、この特定情報は、情報判定機器２が、「特定情報ですよ」と判定して出力するようにしてもよい。ここでは、ステップＳ１における判定結果に基づいて、疾病と関連する特定情報、例えば、健康な状態と差異がある数値などの特徴が検出されたか否かを判定する。 In step S1, when the control unit 1 determines the sensor output result by a determination using a specific program or the like, it then determines whether or not specific information can be obtained (S3). Further, this specific information may be output by the information determination device 2 after determining that it is "specific information". Here, based on the determination result in step S1, it is determined whether or not specific information related to the disease, for example, a feature such as a numerical value different from the healthy state is detected.
　ステップＳ３における判定の結果、特定情報を取得できない場合には、ユーザのプロフィールや行動、生活習慣の判定を行う（Ｓ１５）。特定情報を取得していないタイミングで、端末４を使用しているユーザのプロフィール、行動、生活習慣の判定を行う。ユーザのこれらの情報は、ユーザが端末４から制御部１が通信制御部１ａを通じて入力してもよく、ユーザがＳＮＳ等、ネット上にアップロードしたプロフィール等をダウンロードし、これらの情報を蓄積してもよい。このよう情報の授受は、プッシュ型、プル型、頻繁な取得、間欠的な取得であってもよく、また制御部１が、ユーザの端末４への手入力結果を、通信制御部２を通じて取得し、ＤＢ部８に参照可能に格納してもよい。このＤＢ部８とのやりとりにも通信制御部１ａが介在する。 If specific information cannot be obtained as a result of the determination in step S3, the user's profile, behavior, and lifestyle are determined (S15). The profile, behavior, and lifestyle of the user using the terminal 4 are determined at the timing when the specific information is not acquired. These information of the user may be input by the user from the terminal 4 through the communication control unit 1a by the control unit 1, and the user downloads a profile or the like uploaded on the net such as SNS and accumulates the information. May be good. Such information transfer may be push type, pull type, frequent acquisition, or intermittent acquisition, and the control unit 1 acquires the manual input result of the user to the terminal 4 through the communication control unit 2. However, it may be stored in the DB unit 8 so that it can be referred to. The communication control unit 1a also intervenes in the communication with the DB unit 8.
　これらの情報を蓄積しておくことによって、制御部１が、通信制御部１ａと連携して推論エンジン７を利用し、またＤＢ部８を検索することによって、推奨施設を絞り込む際に（Ｓ１１参照）、適切な情報提供が可能となる。年齢や性別や既往症などの情報や、住所や食習慣や食べ物の情報なども有効である。この情報は端末４にてアンケートを取るような方法、情報判定機器２をセットアップする時に入力して取得する方法、通院時に関連検査機関９にて入力する方法などがあり、これらの装置やその装置を通じてネットワーク上に存在する情報を集めて用意してもよい。 By accumulating this information, when the control unit 1 uses the inference engine 7 in cooperation with the communication control unit 1a and searches the DB unit 8 to narrow down the recommended facilities (see S11). ), Appropriate information can be provided. Information such as age, gender, and pre-existing illness, as well as information on address, eating habits, and food are also effective. This information can be obtained by taking a questionnaire on the terminal 4, inputting and acquiring the information when setting up the information judgment device 2, and inputting by the related inspection institution 9 at the time of going to the hospital. These devices and their devices Information existing on the network may be collected and prepared through.
　ここでは、時系列的に得られた個人ごとの生体情報検査データを利用して、推奨施設等のアドバイスを取得している。さらに、その生体情報検査データの変化パターン情報に対し、改善効果のあった医療機関等の特徴情報をアノテーションした結果、もしくは、特定の健康アドバイスサービスの入力情報をアノテーションした結果を教師データとして学習させた推論モデルを生成し、この推論モデルを利用して検査データの改善時のきっかけ情報を得る場合について説明している。また、変化パターン情報に対し、服薬や生活改善の開始など特定の改善タイミング情報を同様の時間情報として含むことによって、生体情報検査データの時系列パターンが類似である他の人の検査データの改善時のきっかけ情報を特定の人物に伝達する場合、あるいはそれらの組み合わせによる情報伝達方法について説明している。ここでは、検査データの改善時のきっかけ情報は検査や処置のレベルにまで還元した情報であることから、情報を伝達された人（情報を受ける人）の、生活パターンや、生活する地域、地方においても活用可能な情報が伝達されるように変換してアドバイスを提示する工夫を行っている。 Here, advice on recommended facilities, etc. is obtained using the biometric information test data for each individual obtained in chronological order. Furthermore, the result of annotating the characteristic information of medical institutions, etc. that had an improvement effect on the change pattern information of the biometric information test data, or the result of annotating the input information of a specific health advice service is learned as teacher data. The case where the inference model is generated and the trigger information at the time of improvement of the inspection data is obtained by using this inference model is described. In addition, by including specific improvement timing information such as taking medication and starting life improvement as similar time information to the change pattern information, improvement of the test data of other people whose time series pattern of the biometric information test data is similar. Explains how to convey information that triggers time to a specific person, or how to convey information by combining them. Here, since the information that triggered the improvement of the test data is the information that has been reduced to the level of the test or treatment, the life pattern of the person who transmitted the information (the person who receives the information), the area where they live, and the region where they live. We are also devising ways to provide advice by converting information so that it can be used.
　つまり、推奨の絞り込みを行うにあたって、改善効果が何によって得られたかが、単なるどこかの医療機関での治療といった場合には、その医療機関から遠く離れた人には参考にならなかった情報であっても、その医療機関において行われた具体的な検査、診察、処方、服薬、あるいは施設やマンパワーや周辺環境等、その医療機関から離れた地方においても参考に出来るレベルまでかみ砕いた情報（上記推論や上記データベース検索によって得る）にすることが可能である。情報を受ける人の生活に照らし合わせて、条件を満たすもの探すことで、他の地方においても活用可能な情報として取得可能となる。情報を受け取る人の生活圏、もしくはその隣接エリアを条件に加えて、上述の抽出条件を満たす施設等を検索すればよい。もちろん、特定の医療機関が好結果をもたらしていることが顕著であって、情報を受け取る人が近隣の人である場合は、その特定の医療機関に行くアドバイスが提示される。 In other words, when narrowing down the recommendations, what was the improvement effect was information that was not helpful to people far away from that medical institution when it was just treatment at some medical institution. However, specific tests, medical examinations, prescriptions, medications, facilities, manpower, surrounding environment, etc. performed at the medical institution, etc., have been chewed to a level that can be referred to even in regions away from the medical institution (above). It can be obtained by inference or the above database search). By searching for information that meets the conditions in light of the lives of the people who receive the information, it will be possible to obtain information that can be used in other regions as well. In addition to the living area of the person who receives the information or the area adjacent to the living area, a facility or the like that satisfies the above extraction conditions may be searched. Of course, if it is noticeable that a particular medical institution is producing good results and the person receiving the information is a neighbor, advice is given to go to that particular medical institution.
　ステップＳ３における判定の結果、特定情報を取得できた場合には、データベースが構築されているか否かを判定する（Ｓ５）。ステップＳ３において取得されたと判定された特定情報を用いて検索・推論するに適したデータベースがＤＢ部８に蓄積されているか否かを、制御部１が通信制御部１ａと連携して判定する。例えば、ステップＳ３における判定の結果、大便の色を検査した結果、赤色を表す数値が健康の場合に比較して大きくなった場合に、このような健康状態を判定するに適したデータベースがＤＢ部８に蓄積されているか否かを判定する。なお、制御部１を含む情報伝達システム内のＤＢ部８には、適したデータベースが蓄積されていなくても、他のシステム内に存在する場合がある。そこで、他のシステムを含めてデータベースを検索するようにしてもよい。 If the specific information can be acquired as a result of the determination in step S3, it is determined whether or not the database has been constructed (S5). The control unit 1 determines in cooperation with the communication control unit 1a whether or not a database suitable for searching / inferring using the specific information determined in step S3 is stored in the DB unit 8. For example, as a result of the determination in step S3, as a result of inspecting the color of the stool, when the numerical value representing red becomes larger than that in the case of health, a database suitable for determining such a health condition is stored in the DB section. It is determined whether or not it is accumulated in 8. Even if a suitable database is not stored in the DB unit 8 in the information transmission system including the control unit 1, it may exist in another system. Therefore, the database may be searched including other systems.
　ステップＳ５における判定の結果、制御部１が通信制御部１ａと連携してデータベースが有ると判断した場合には、制御部１が通信制御部１ａと連携し、ＤＢ部８に記録されている保有設備を含め、関連施設情報を取得する（Ｓ７）。ここでは、特定の患者候補の上記情報判定機器２が出力するデータ（検査データ、生体データ、バイタルデータ、検体データなどを総じて「時系列データ」として表記）がどのような時間的変化や推移をしているかを、パターン照合等のロジックベースで類似パターンの人をデータベース（ＤＢ部８）から探し、その中で特定の医療機関に受診している人を探して、その医療機関の特徴を抽出すればよい。 As a result of the determination in step S5, when the control unit 1 determines that the database exists in cooperation with the communication control unit 1a, the control unit 1 cooperates with the communication control unit 1a and is recorded in the DB unit 8. Acquire related facility information including equipment (S7). Here, what kind of temporal changes and transitions are made in the data output by the above-mentioned information determination device 2 of a specific patient candidate (test data, biological data, vital data, sample data, etc. are generally expressed as "time series data"). Search for people with similar patterns from the database (DB section 8) based on logic such as pattern matching, search for people who are consulting at a specific medical institution, and extract the characteristics of that medical institution. do it.
　医療機関の受診の前であれば。類似パターンに基づいて、医療機関受診にともなう服薬、生活改善などによって、時系列データパターンの傾向が良好に転じたものを探す方が良く、患者候補も同様の行動を起こすことによって、生体データパターンを良好な方向にできる可能性がある。したがって、これは単なるデータのパターン判定ではなく、以下のような特徴を持ったパターン判定になっている。つまり、特定の仕様の検査機器を用いて時系列的に検査データを取得し、かつ、この時系列パターンの時間軸に合わせて、服薬や生活改善の開始など特定の改善タイミング情報を有するデータベースを使った判定であり、これは、特定の改善タイミング以降、該検査データの変化パターン情報も取得してあって、変化パターンが確かに改善されているという特徴が表されている例をも含みうるデータベースになっているため、当該改善タイミング以前において、将来の患者の変化パターンの類似性を判定することが好ましい。改善タイミング以降に行った、何らかの措置の情報（これは時系列パターンと同じ時間情報に関連付けられた情報となっている）こそが、状況改善のヒント情報になるからである。 Before seeing a medical institution. Based on a similar pattern, it is better to look for a time-series data pattern whose tendency has changed satisfactorily due to medication, lifestyle improvement, etc. associated with medical institution consultation, and patient candidates also take similar actions to obtain biometric data patterns. May be in the right direction. Therefore, this is not a mere pattern determination of data, but a pattern determination having the following characteristics. In other words, a database that acquires test data in chronological order using a test device with specific specifications and has specific improvement timing information such as taking medication and starting life improvement according to the time axis of this time series pattern. It is a judgment used, and this may include an example in which the change pattern information of the inspection data is also acquired after a specific improvement timing, and the feature that the change pattern is certainly improved is shown. Since it is a database, it is preferable to determine the similarity of future patient change patterns before the improvement timing. This is because the information on some measures taken after the improvement timing (which is the information associated with the same time information as the time series pattern) is the hint information for improving the situation.
　言い換えると、上述のデータベースは、特定の仕様の検査機器を用いて時系列的に検査データを取得し、該検査データの変化パターン情報を取得し、この変化パターン情報に対し、服薬や生活改善の開始など特定の改善タイミング情報を変化パターン情報の時間情報と同様の時間情報として含むことによって、当該改善情報以前の変化パターンの部分を比較可能にした情報を有している。また、特定の人物の変化パターンの部分を他の人の改善情報以前の変化パターンの部分と比較して、改善情報を特定の人物に伝達することを特徴とする情報伝達方法が提供可能となる。 In other words, the above-mentioned database acquires inspection data in chronological order using an inspection device having a specific specification, acquires change pattern information of the inspection data, and responds to this change pattern information by taking medication or improving life. By including specific improvement timing information such as start as time information similar to the time information of the change pattern information, it has information that makes it possible to compare the part of the change pattern before the improvement information. Further, it becomes possible to provide an information transmission method characterized by transmitting improvement information to a specific person by comparing a part of a change pattern of a specific person with a part of a change pattern before the improvement information of another person. ..
　別の言い方をすると、上述のデータベースは、特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得して作成した検査データの変化パターン情報と、特定の期間内における上記検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録している。このデータベースを用いて、変化パターンのトレンド変化とイベントのタイミングの関係に基づいて、イベント情報のうち、トレンド変化に影響したイベントを抽出して情報提供を可能としている。 In other words, the above-mentioned database is created by acquiring a plurality of inspection data in time series using an inspection device having a specific specification and a specific inspection data change pattern information. Each timing of a plurality of event information regarding the acquisition source of the inspection data within the period is recorded in association with each other. Using this database, it is possible to extract and provide information on events that have influenced the trend change from the event information based on the relationship between the trend change of the change pattern and the timing of the event.
　上述したような工夫によって、ステップＳ５においては、ステップＳ３において取得した特定情報から疑われる特定の疾患を検査、治療するにふさわしい関連施設情報や生活習慣情報を取得することが可能となる。このとき、特定の疾患を検査、治療するために使用する設備を保有する関連施設情報を取得し、この施設情報を候補とする。例えば、前述の大便が赤色化している場合には、痔疾や大腸癌が疑われるが、色の特徴や時系列的変化の特徴に基づいて更に詳細な分類が可能であることから、例えば、肛門科ではなく、大腸内視鏡を有し大腸内視鏡検査が可能な施設に関する情報を選り分けて取得することが可能となる。 By devising as described above, in step S5, it becomes possible to acquire related facility information and lifestyle information suitable for examining and treating a specific disease suspected from the specific information acquired in step S3. At this time, the related facility information possessing the equipment used for inspecting and treating a specific disease is acquired, and this facility information is used as a candidate. For example, if the above-mentioned stool is red, hemorrhoids or colon cancer are suspected, but since more detailed classification is possible based on the characteristics of color and the characteristics of time-series changes, for example, the anus. It will be possible to sort out and obtain information about facilities that have colonoscopies and can perform colonoscopy instead of departments.
　ステップＳ５における判定の結果、データベースがない場合には、履歴データを用いて推論する（Ｓ９）。ステップＳ３において取得した特定情報に適したデータベースが用意されていない場合もあり得る。この場合には、履歴データを推論エンジン７に入力し、推論エンジン７が推論モデルを用いて、適切な検査が可能な施設等の推論を行う。これは、特定情報と関係のある疾患を推論し、また特定情報と関係のある診療科・部門を推論する。そして、この推論結果に基づいて、ユーザの健康状態に相応しい検査施設（機関）・診療施設（機関）を推奨する。また、このステップでは、推奨する医療施設の推論以外にも、ユーザの病名の推論や、ユーザの症状が悪くなり医院に治療を受けに行く時期の推論等も行う。この履歴データを用いた推論の詳しい動作については、図５を用いて後述する。 If there is no database as a result of the determination in step S5, inference is made using historical data (S9). It is possible that a database suitable for the specific information acquired in step S3 is not prepared. In this case, the historical data is input to the inference engine 7, and the inference engine 7 uses the inference model to infer the facilities and the like capable of appropriate inspection. This infers diseases related to specific information, and infers clinical departments / departments related to specific information. Then, based on this inference result, an examination facility (institution) / medical care facility (institution) suitable for the user's health condition is recommended. In addition to inferring the recommended medical facility, this step also infers the name of the user's disease and the time when the user's symptoms worsen and go to the clinic for treatment. The detailed operation of inference using this historical data will be described later with reference to FIG.
　ステップＳ７において、関連施設情報を取得すると、またはステップＳ９において、推奨施設を推論すると、次に、ユーザのプロフィール等に基づいて、推奨施設を絞り込む（Ｓ１１）。ここでは、ステップＳ１５において取得したプロフィール情報や、関連検査機関９から取得した情報、例えば、ユーザの近くのクリニック情報、ユーザの休日に診療を行っている病院情報に基づいて、ユーザが通院するに便利な施設を選んでもよい。この選択は、制御部１が検索部１ｆと連携してＤＢ部８に蓄積された情報から条件に適ったものの取得することによって行う。 When the related facility information is acquired in step S7 or the recommended facility is inferred in step S9, then the recommended facility is narrowed down based on the user's profile and the like (S11). Here, the user goes to the hospital based on the profile information acquired in step S15, the information acquired from the related inspection institution 9, for example, the clinic information near the user, and the hospital information on the user's holiday. You may choose a convenient facility. This selection is performed by the control unit 1 cooperating with the search unit 1f to acquire information suitable for the conditions from the information stored in the DB unit 8.
　次に、制御部１が、通信制御部１ａと連携して、ユーザの端末４に推奨施設を表示する（Ｓ１３）。ここでは、特定情報を提供したユーザに対して、ステップＳ１１において絞り込んだ推奨施設を、表示する。すなわち、ステップＳ１３は、ユーザやその関係者に、検査や診療補助の情報を提供するステップであり、端末４に表示する。また、ユーザの健康状態によっては、警告表示を行ってもよい。 Next, the control unit 1 cooperates with the communication control unit 1a to display the recommended facility on the user's terminal 4 (S13). Here, the recommended facilities narrowed down in step S11 are displayed for the user who provided the specific information. That is, step S13 is a step of providing information on examinations and medical assistance to users and related persons, and is displayed on the terminal 4. Further, depending on the health condition of the user, a warning display may be displayed.
　このように、本実施形態における検査結果送信のフローにおいては、制御部１は、情報判定機器２等からのセンサ検出結果を取得し（Ｓ１）、これらの検出結果から健康状態（疾病）に関係する特定情報があるか否かを判定している（Ｓ３）。特定情報が有った場合には、データベースを検索し、健康状態（疾病）を検査・診察するに適した施設を、保有設備を含めて検索している（Ｓ７）。このため、日々、日常生活の中で、ユーザは健康チェックを行うことができ、また健康状態に応じて、施設の保有設備を考慮し、検査・診察に適した施設のアドバイスを受けることができる。 As described above, in the flow of transmitting the inspection result in the present embodiment, the control unit 1 acquires the sensor detection results from the information determination device 2 and the like (S1), and is related to the health state (disease) from these detection results. It is determined whether or not there is specific information to be used (S3). When there is specific information, the database is searched to search for facilities suitable for examining and diagnosing the health condition (disease), including the owned equipment (S7). For this reason, the user can perform health checks in daily life, and can receive advice on facilities suitable for examinations and medical examinations, considering the facilities owned by the facilities according to the health condition. ..
　なお、図４に示すフローチャートでは、前述したように、ＤＢ検索（Ｓ７）と推論（Ｓ９）を別の処理として独立に扱った。しかし、これに限らず、これらを総合的に扱ってもよい。例えば、推論を行った後、ＤＢを検索するような方法もあり、学習時に検査装置情報を含むＤＢ内の情報も含めて学習した推論モデルを使用して、保有する機材等の設備まで出力する推論を行ってもよい。この場合には、「このクリニックには〇〇検査装置があります」といった表示が可能になる。 In the flowchart shown in FIG. 4, as described above, the DB search (S7) and the inference (S9) are treated independently as separate processes. However, the present invention is not limited to this, and these may be treated comprehensively. For example, there is a method of searching the DB after making an inference, and using the inference model learned including the information in the DB including the inspection device information at the time of learning, the equipment such as the equipment owned is output. You may make inferences. In this case, it is possible to display such as "This clinic has XX inspection equipment".
　次に、図５に示すフローチャートを用いて、図４のステップＳ９の「履歴データを用いて推論」の動作について説明する。このフローは、前述したように、予め定められた時間幅で対象者の検査データの変化パターンを抽出し、この変化パターンに基づいて、推奨施設を推論する。制御部１が、通信制御部１ａを通じて、推論エンジン７やＤＢ部８などと連携して行う。 Next, the operation of "inference using historical data" in step S9 of FIG. 4 will be described using the flowchart shown in FIG. In this flow, as described above, the change pattern of the test data of the subject is extracted within a predetermined time width, and the recommended facility is inferred based on this change pattern. The control unit 1 performs the operation in cooperation with the inference engine 7, the DB unit 8, and the like through the communication control unit 1a.
　図５に示すフローが開始すると、時系列データを取得する（Ｓ２１）。ここでは、ＤＢ部８に記録されている特定ＩＤに対応する時系列のデータを取得する（時系列データについては、図３を参照）。取得する時系列データの時間幅は、特定の時間幅にするが、特定の時間幅のデータを取得できない場合には、取得できる時間範囲とする。特定の時間幅がないと、特定の状況下のデータのみでの判定となり信頼性が劣るからである。特定の時間幅は、大腸癌のように時間をかけて進行する疾病と、インフルエンザのように短期間で進行する疾病では、異なる。また、検査データの種別は、推論モデルの学習に依存するが、学習時に使用した特定の項目のグラフになることが望ましく、例えば体重と血圧を一緒に推論しない事が好ましい。 When the flow shown in FIG. 5 starts, time series data is acquired (S21). Here, the time-series data corresponding to the specific ID recorded in the DB unit 8 is acquired (see FIG. 3 for the time-series data). The time width of the time-series data to be acquired is set to a specific time width, but if the data of the specific time width cannot be acquired, it is set to the time range that can be acquired. This is because if there is no specific time width, the judgment will be based only on the data under a specific situation, and the reliability will be inferior. The specific time span differs between diseases that progress over time, such as colorectal cancer, and diseases that progress in a short period of time, such as influenza. Further, the type of test data depends on the learning of the inference model, but it is desirable that the graph is a specific item used at the time of learning. For example, it is preferable not to infer body weight and blood pressure together.
　時系列データを取得すると、次に、変化が顕著か否かを判定する（Ｓ２３）。ここでは、取得した時系列データの変化パターンが顕著か否かを判定する。この判定の結果、変化が顕著な場合には、警告を行う（Ｓ２５）。特定情報の値そのものが標準値から大きく外れている場合や、変化パターンが顕著な場合には、何らかの疾病がある可能性が高いことから、ステップＳ２７の処理を実行することなく、ユーザに警告表示する。この警告表示は、例えば、端末４に表示してもよい。警告表示を行うと、このフローを終了し、元のフローに戻る。 After acquiring the time series data, it is next determined whether or not the change is remarkable (S23). Here, it is determined whether or not the change pattern of the acquired time series data is remarkable. If the result of this determination is significant, a warning is given (S25). If the value of the specific information itself deviates significantly from the standard value, or if the change pattern is remarkable, there is a high possibility that there is some kind of disease, so a warning is displayed to the user without executing the process of step S27. To do. This warning display may be displayed on the terminal 4, for example. When a warning is displayed, this flow is terminated and the original flow is restored.
　一方、ステップＳ２３における判定の結果、変化が顕著でない場合には、次に、問題のないレベルか否かを判定する（Ｓ２７）。ここでは、ステップＳ２１において取得した時系列データの変化パターンが問題のない程度のレベルであるか否かを判定する。図３（ｃ）に示したように、特定の時間幅でみた場合に、検査データのレベルが所定レベル以下であれば、問題がないレベルと判定する。このステップＳ２７における判定の結果、変化が問題ないレベルであれば、医療機関情報を出力しない（Ｓ２９）。特に、ユーザに表示すべき情報がないことから、医療機関に関する情報を出力しない。ステップＳ２９を実行すると、元のフローに戻る。 On the other hand, if the change is not remarkable as a result of the determination in step S23, then it is determined whether or not the level has no problem (S27). Here, it is determined whether or not the change pattern of the time series data acquired in step S21 is at a level at which there is no problem. As shown in FIG. 3C, if the level of the inspection data is equal to or lower than the predetermined level when viewed in a specific time width, it is determined that there is no problem. As a result of the determination in step S27, if there is no problem in the change, the medical institution information is not output (S29). In particular, since there is no information to be displayed to the user, information about the medical institution is not output. When step S29 is executed, the flow returns to the original flow.
　ステップＳ２７における判定の結果、問題のあるレベルであれば、次に、特定時間幅分の時系列データを取得できたか否かを判定する（Ｓ３１）。例えば、潜血の状況を検出したのであれば、潜血が数か月の幅で得られたか否かを判定する。すなわち、特定の時間幅は、関連する疾病に応じて異なる。 If the result of the determination in step S27 is a problematic level, then it is determined whether or not the time series data for a specific time width can be acquired (S31). For example, if the status of occult blood is detected, it is determined whether or not occult blood was obtained within a range of several months. That is, the specific time span depends on the disease involved.
　ステップＳ３１における判定の結果、特定の時間幅のデータを取得できた場合には、データを推論部に入力させ、推論結果を取得する（Ｓ３５）。特定の時間幅のデータを取得できた場合だけ、推論エンジン７にデータを入力し、推論結果を得ている。同様のデータを集めて学習した推論モデルであれば、「このような傾向の人は、すでに通院している人がいる」とか、「何か月後に通院することになる」、といった推論結果を得ることができる。学習部５が、推奨設備まで含めて、「〇〇の検査装置がある医院に行く人が多い」といった推論結果を出力できる推論モデルを、生成する学習も可能である。図３を用いて説明したように、時系列データに基づいて教師データを生成することによって、通院が必要になるか否か、通院することになる場合にはその時期や、どの診療科に通院することになるか等、種々の推論モデルを生成することができる。 If the data of a specific time width can be acquired as a result of the determination in step S31, the data is input to the inference unit and the inference result is acquired (S35). Only when the data of a specific time width can be acquired, the data is input to the inference engine 7 and the inference result is obtained. With an inference model learned by collecting similar data, inference results such as "some people with this tendency have already gone to the hospital" or "will go to the hospital in a few months" can be obtained. Obtainable. It is also possible for the learning unit 5 to generate an inference model that can output an inference result such as "many people go to a clinic with a XX inspection device" including recommended equipment. As explained with reference to FIG. 3, by generating teacher data based on time-series data, whether or not it is necessary to go to the hospital, when it is necessary to go to the hospital, and which clinical department to go to. Various inference models can be generated, such as whether to do so.
　また、ステップＳ３５においても、図２のように整理されたデータベースを利用している（生体データ変化パターンのアノテーションに利用されている）ので、単純に推論結果として医療機関が出力されるのではなく、その施設の様々な特徴などが推論結果として出力され、個々の医療機関に依存しない結果が得られる。なお、学習に使われる生体データ変化パターンは、すでに通院している人のデータを教師データとする時、通院、服薬等で改善される前のデータを選択して学習することによって、診療を受けるかどか迷っている人や、自覚症状もなく診療を受けていない人の生体データパターンとの比較がしやすくなる。 Further, also in step S35, since the database organized as shown in FIG. 2 is used (used for annotation of the biological data change pattern), the medical institution is not simply output as the inference result. , Various features of the facility are output as inference results, and results that do not depend on individual medical institutions can be obtained. In addition, the biological data change pattern used for learning receives medical treatment by selecting and learning the data before it is improved by going to the hospital, taking medication, etc. when the data of the person who has already visited the hospital is used as the teacher data. It will be easier to compare with the biometric data patterns of those who are wondering whether or not they are, or those who have no subjective symptoms and are not receiving medical treatment.
　したがって、本願の実施形態におけるデータベースは、特定の仕様の検査機器を用いて時系列的変化パターン情報を取得し、変化パターン情報に対し、服薬や生活改善の開始など特定の改善タイミング情報を同様の時間情報として含むことによって、当該改善情報以前の変化パターンを教師データ化可能にした情報を有している。また、本願の実施形態における情報伝達方法は、特定の人物の変化パターンを他の人の改善情報以前の変化パターンを入力として、推論結果として得られた改善情報を特定の人物に伝達するようにしている。 Therefore, the database according to the embodiment of the present application acquires time-series change pattern information using an inspection device having a specific specification, and similarly performs specific improvement timing information such as taking medication or starting life improvement with respect to the change pattern information. By including it as time information, it has information that makes it possible to convert the change pattern before the improvement information into teacher data. Further, in the information transmission method in the embodiment of the present application, the change pattern of a specific person is input to the change pattern before the improvement information of another person, and the improvement information obtained as an inference result is transmitted to the specific person. ing.
　ステップＳ３１における判定の結果、特定時間幅のデータを取得していない場合には、推論を行わない（Ｓ３３）。特定の時間幅の情報がなくても、期待する信頼性によっては推論が可能であるが、それも難しい場合がある。そこで、ステップＳ３１において、特定時間幅のデータを取得していないと判定された場合には、推論しない。ただし、明らかに危険な状況を検出できる場合もあり、その場合は、推論以前に、緊急情報を出力すれば良い。 As a result of the determination in step S31, if the data of the specific time width has not been acquired, no inference is performed (S33). Inference is possible depending on the expected reliability without information of a specific time width, but it can be difficult. Therefore, if it is determined in step S31 that the data of the specific time width has not been acquired, no inference is made. However, there are cases where a clearly dangerous situation can be detected, in which case emergency information may be output before inference.
　つまり、取得したデータが、顕著に問題のある数値の場合には、推論で長期予想をする時間的猶予はなく、ステップＳ２３、Ｓ２５において対応する。この対応によってステップＳ３３において推論をしなくても緊急時に対応できないという問題を防止し、十分なデータが集まってから、情報を出すという信頼性の高いシステムにすることが出来る。つまり、本実施形態では、特定の変化で収まっている数値変化の場合、あらかじめ定められた時間幅で対象者の検査データの変化パターンを切り取って時間情報と共に学習された推論モデルに従って推論を行う。推論結果を得ると、このフローを終了し、元のフローに戻る。 In other words, if the acquired data is a numerical value that has a significant problem, there is no time grace to make a long-term forecast by inference, and it will be dealt with in steps S23 and S25. This response prevents the problem of not being able to respond to an emergency without making inferences in step S33, and makes it possible to create a highly reliable system in which information is output after sufficient data has been collected. That is, in the present embodiment, in the case of a numerical change that is contained in a specific change, the change pattern of the test data of the subject is cut out within a predetermined time width, and inference is performed according to the inference model learned together with the time information. When the inference result is obtained, this flow is terminated and the original flow is restored.
　このように、図５に示す履歴データを用いて推論のフローでは、時系列データを取得すると（Ｓ２１）、この時系列データの変化が顕著な場合には警告表示を行い（Ｓ２５）、特定時間幅分の時系列データを取得できた場合には（Ｓ３１Ｙｅｓ）、履歴データを用いて推論を行っている（Ｓ３５）。本フローでは、特定時間幅の時系列データを用いて、推論を行っていることから、精度の高い推論を行うことができる。 As described above, in the inference flow using the historical data shown in FIG. 5, when the time series data is acquired (S21), a warning is displayed when the change of the time series data is remarkable (S25), and the specific time is specified. When the time series data for the width can be acquired (S31Yes), the inference is performed using the historical data (S35). In this flow, inference is performed using time-series data having a specific time width, so that inference can be performed with high accuracy.
　また、時系列データのレベルが問題ないレベルか否かを判定し、問題なしのレベルであれば、情報を出力することがない（Ｓ２９）。このように、時系列データが問題なしのレベルであれば、余計な推論を行わせないために、情報の出力を控えもよい。つまり、その生体情報変化とは別の原因の疾病がある場合、この情報が教師データに入ってしまった場合に、関係ない推論結果を出力する可能性を排除している。つまり、本実施形態においては、特定の基準を満たさない場合に（例えば、Ｓ２３Ｙｅｓ、Ｓ２７Ｙｅｓ）、予め定められた時間幅で対象者の検査データの変化パターンを切り取って時間情報と共に学習された推論モデルに従って推論は行わない。人が目で見て検出できないような微妙な変化から、何かを推論することに優れている学習の効果を期待し、また、あまりにも微妙な推論を行う場合の弊害を排除している。 In addition, it is determined whether or not the level of the time series data is at a level that does not cause any problem, and if there is no problem, no information is output (S29). In this way, if the time series data is at a level where there is no problem, it is possible to refrain from outputting the information in order to prevent unnecessary inference. That is, when there is a disease having a cause other than the change in biological information, the possibility of outputting an unrelated inference result when this information is included in the teacher data is excluded. That is, in the present embodiment, when the specific criteria are not satisfied (for example, S23Yes, S27Yes), the inference model learned together with the time information by cutting out the change pattern of the inspection data of the subject within a predetermined time width. No inference is made according to. We expect the learning effect to be excellent in inferring something from subtle changes that cannot be visually detected by humans, and eliminate the harmful effects of making too subtle inferences.
　なお、図５に示すフローにおいては、推論エンジン７による推論は、ステップＳ３５において実行している。しかし、ステップＳ２３、Ｓ２７における判定を推論エンジン７によって行ってもよい。この場合、同一の推論モデルを用いて実行してもよく、またそれぞれ別々の推論エンジンで実行してもよい。 In the flow shown in FIG. 5, the inference by the inference engine 7 is executed in step S35. However, the determination in steps S23 and S27 may be made by the inference engine 7. In this case, it may be executed using the same inference model, or it may be executed by different inference engines.
　次に、図６に示すフローチャートを用いて、図４のステップＳ７の「関連施設情報の取得」の動作の変形例を説明する。図４のステップＳ７においては、単純にデータベースを検索していた。この検索には、取得した時系列データが、特定の診療科と関連している必要があったが、実際には、特定の診療科に関係しない情報が必要な場合がある。例えば、体重の増加や血圧の上昇などは、特定の診療科や特定の装置以外の検査を要することがある。図６に示す変形例は、このような場合にも適切に関連施設情報を取得することができる。 Next, using the flowchart shown in FIG. 6, a modified example of the operation of “acquisition of related facility information” in step S7 of FIG. 4 will be described. In step S7 of FIG. 4, the database was simply searched. For this search, the acquired time-series data had to be related to a specific clinical department, but in reality, information not related to a specific clinical department may be required. For example, gaining weight or increasing blood pressure may require tests other than specific departments or specific devices. In the modified example shown in FIG. 6, related facility information can be appropriately acquired even in such a case.
　制御部１が通信制御部１ａと連携して図６に示すフローの動作を開始すると、まず、特定診療科用の情報であるか否かを判定する（Ｓ４１）。ここでは、制御部１が、取得する時系列データが特定診療科用の情報であったか否かを判定する。 When the control unit 1 starts the operation of the flow shown in FIG. 6 in cooperation with the communication control unit 1a, it first determines whether or not the information is for a specific clinical department (S41). Here, the control unit 1 determines whether or not the time-series data to be acquired is information for a specific clinical department.
　ステップＳ４１における判定の結果、特定診療科用でなかった場合には、ステップＳ４３～Ｓ４７において、ＤＢ部８に記録されているデータに加え、複数の情報（データ）の遷移を用いて、適切な診療科を出力するような推論モデルを使うことを考慮している。つまり、本実施形態においては、特定の変化で収まっている数値変化の場合、制御部１は、予め定められた時間幅で対象者の検査データの変化パターンを抽出し、時間と共に変化するデータを用いて学習することによって推論モデルを生成し、この推論モデルを用いて推論を行い、その推論結果と保有機器データベースを参照することによって情報を出力する。ここで使用する推論モデルは、学習時に図３（ａ）～図３（ｃ）に示したような時刻情報付きデータ（図２のＤｙ（ｔ）参照）と、そのユーザが受ける診療科を入出力として学習したものを想定している。 As a result of the determination in step S41, if it is not for a specific clinical department, in steps S43 to S47, in addition to the data recorded in the DB unit 8, a plurality of information (data) transitions are used to be appropriate. We are considering using an inference model that outputs the clinical department. That is, in the present embodiment, in the case of a numerical change that is contained in a specific change, the control unit 1 extracts a change pattern of the inspection data of the subject within a predetermined time width, and extracts the data that changes with time. An inference model is generated by learning using this inference model, inference is performed using this inference model, and information is output by referring to the inference result and the possessed device database. The inference model used here includes data with time information (see Dy (t) in FIG. 2) as shown in FIGS. 3 (a) to 3 (c) during learning and the clinical department received by the user. It is assumed that it is learned as an output.
　ステップＳ４１における判定の結果、特定診療科用でなかった場合には、まず、時系列データを取得する（Ｓ４３）。ここでは、制御部１が情報判定機器２等からデータの過去分を履歴として取得する。これらのデータは、制御部１内のメモリまたはＤＢ部８に記録されている。 As a result of the determination in step S41, if it is not for a specific clinical department, first, time series data is acquired (S43). Here, the control unit 1 acquires the past data from the information determination device 2 and the like as a history. These data are recorded in the memory in the control unit 1 or the DB unit 8.
　続いて、この取得した時系列データが、特定時間幅分のデータであるか否かを判定する（Ｓ４５）。この判定の結果、特定時間幅分のデータがない場合には、一般クリニックを推奨する（Ｓ５３）。時系列データの時間幅が、特定時間ないことから、関連疾病を推論できず、特定の施設を紹介できないため、一般的な医療施設を推奨する。なお、図５のＳ２９のように、医療機関情報を出力しないとしてもよい。 Subsequently, it is determined whether or not the acquired time series data is data for a specific time width (S45). As a result of this determination, if there is no data for a specific time width, a general clinic is recommended (S53). Since the time width of the time series data does not have a specific time, it is not possible to infer related diseases and refer to a specific facility, so a general medical facility is recommended. Note that the medical institution information may not be output as in S29 of FIG.
　ステップＳ４５における判定の結果、特定時間幅分のデータを取得できた場合には、データを推論部に入力し、関連疾病情報を取得する（Ｓ４７）。ここでは、制御部１は、取得したデータを推論エンジン７に入力し、関連疾病情報の推論を行う。この時使用する推論モデルは、図３に示したような情報を教師データとして生成する。 If the data for a specific time width can be acquired as a result of the determination in step S45, the data is input to the inference unit and the related disease information is acquired (S47). Here, the control unit 1 inputs the acquired data to the inference engine 7 and infers the related disease information. The inference model used at this time generates information as shown in FIG. 3 as teacher data.
　ステップＳ４７において関連疾病情報を取得すると、またはステップＳ４１における判定の結果、特定診療科用の情報であった場合には、次に、関連設備を判定し（Ｓ４９）、設備保有施設情報（診療科）を取得する（Ｓ５１）。制御部１は、ステップＳ４１において判定された特定診療科、またはステップＳ４７において取得した疾病に関連した設備が何であるかを判定する。そして、制御部１は、判定された設備を有する施設をＤＢ部８の施設別保有機器一覧８ａから検索し、選択する。 If the related disease information is acquired in step S47, or as a result of the determination in step S41, the information is for a specific clinical department, then the related equipment is determined (S49), and the equipment possession facility information (clinical department) is determined. ) Is acquired (S51). The control unit 1 determines what the specific clinical department determined in step S41 or the equipment related to the disease acquired in step S47 is. Then, the control unit 1 searches for and selects a facility having the determined equipment from the facility-specific equipment list 8a of the DB unit 8.
　次に、図７に示すフローチャートを用いて、情報伝達システムにおける検査結果の送信の変形例の動作について説明する。このフローも図４と同様に、主として、制御部１内のＣＰＵがメモリに記憶されたプログラムに従って、情報伝達システム全体を制御することによって、実行される。また、図４に示すフローは、図１に示したＤＢ部８においてロジックベースによる類似パターン判定による検索と、推論エンジン７による推論等の機能を併用していたが、図７に示す例では、推論エンジンを単独で利用する場合を示す。なお、図７においても、推論エンジン単独ではなく、データベースの検索機能も併用するようにしてもよい。また、データべースそのものは、同様に図２のようなものを想定し、ここに記録されている多くの患者の情報を利用した学習によって得られた推論モデルを使用した推論を想定している。図７に示すフローチャートは、図４に示すフローチャートにおけるステップＳ５～Ｓ１３を、ステップＳ４～Ｓ１３ａに置き換えた点において相違するので、この相違点を中心に説明する。 Next, using the flowchart shown in FIG. 7, the operation of a modified example of transmission of inspection results in the information transmission system will be described. Similar to FIG. 4, this flow is also mainly executed by the CPU in the control unit 1 controlling the entire information transmission system according to the program stored in the memory. Further, in the flow shown in FIG. 4, in the DB unit 8 shown in FIG. 1, functions such as a search based on a logic-based similarity pattern determination and an inference by an inference engine 7 are used in combination. The case where the inference engine is used alone is shown. In FIG. 7, not only the inference engine alone but also the database search function may be used together. Similarly, the database itself is assumed to be as shown in FIG. 2, and inference using an inference model obtained by learning using the information of many patients recorded here is assumed. There is. Since the flowchart shown in FIG. 7 is different in that steps S5 to S13 in the flowchart shown in FIG. 4 are replaced with steps S4 to S13a, this difference will be mainly described.
　図７に示す検査結果送信のフローが開始すると、まず、ＩＤ毎に、センサ出力結果に基づいて判定する（Ｓ１）。ここでは、図４のフローと同様に、制御部１がセンサ出力結果に添付されたＩＤ毎に、センサ出力に基づいて、検査結果の判定を行う。続いて、特定情報を得ることが出来たか否かを判定する（Ｓ３）。ここでは、図４のフローと同様に、制御部１がステップＳ１における判定に基づいて、疾病と関連する特定情報、例えば、健康な状態と差異がある数値などの特徴が検出されたか否かを判定する。ステップＳ３における判定の結果、特定情報を取得できない場合には、図４のフローと同様に、ユーザのプロフィールや行動、生活習慣の判定を行う（Ｓ１５）。この処理を実行すると、ステップＳ１に戻る。 When the flow of inspection result transmission shown in FIG. 7 starts, first, each ID is determined based on the sensor output result (S1). Here, as in the flow of FIG. 4, the control unit 1 determines the inspection result for each ID attached to the sensor output result based on the sensor output. Subsequently, it is determined whether or not the specific information can be obtained (S3). Here, as in the flow of FIG. 4, whether or not the control unit 1 has detected specific information related to the disease, for example, a feature such as a numerical value different from the healthy state, based on the determination in step S1. judge. If the specific information cannot be obtained as a result of the determination in step S3, the user's profile, behavior, and lifestyle are determined in the same manner as in the flow of FIG. 4 (S15). When this process is executed, the process returns to step S1.
　ステップＳ３における判定の結果、特定情報を取得できた場合には、推論モデルが有るか否かを判定する（Ｓ４）。制御部１内の推論モデル仕様決定部１ｄは、対象者にアドバイス（推奨施設の表示等）するための推論モデルの仕様を決定し、推論依頼部１ｅは学習部５に対して推論モデルの生成を依頼する。学習部５によって生成された推論モデルを受信すると、この推論モデル７に設定している。推論モデルは、種々の状況に対応できるように複数作成され、制御部１内または推論エンジン７内に記憶されている。このステップ４においては、ステップＳ３において取得した特定情報に対して、対象者に適切なアドバイス（推奨施設の表示等）することが可能な推論モデルが制御部１内等に格納されているか否かを判定する。 If specific information can be obtained as a result of the determination in step S3, it is determined whether or not there is an inference model (S4). The inference model specification determination unit 1d in the control unit 1 determines the inference model specifications for giving advice (display of recommended facilities, etc.) to the target person, and the inference request unit 1e generates an inference model for the learning unit 5. To ask. When the inference model generated by the learning unit 5 is received, it is set in the inference model 7. A plurality of inference models are created so as to correspond to various situations, and are stored in the control unit 1 or the inference engine 7. In this step 4, whether or not an inference model capable of giving appropriate advice (display of recommended facilities, etc.) to the target person for the specific information acquired in step S3 is stored in the control unit 1 or the like. To judge.
　ステップＳ４における判定の結果、推論モデルが有った場合には、履歴データを用いて推論を行う（Ｓ６）。ここでは、ステップＳ４において選択された推論モデルを推論エンジン７に設定する。そして、情報判定機器２から送信され、ＤＢ８に記録されている対象者の履歴データ（図２の時系列データ参照）を、推論エンジン７の入力層に入力し、推論を行うと、推論エンジンの出力層から推論結果が出力される。なお、ステップ６における推論は、図５に示したフローを実行しても勿論かまわない。 If there is an inference model as a result of the determination in step S4, inference is performed using the historical data (S6). Here, the inference model selected in step S4 is set in the inference engine 7. Then, the historical data of the target person (see the time-series data in FIG. 2) transmitted from the information determination device 2 and recorded in the DB 8 is input to the input layer of the inference engine 7, and inference is performed. The inference result is output from the output layer. Of course, the inference in step 6 may be performed by executing the flow shown in FIG.
　ステップ６において推論結果を取得すると、次に、この推論結果の信頼性が高いか否かを判定する（Ｓ８）。ここでは、ステップＳ６における推論結果の信頼性を算出し、この信頼性が所定値よりも高いか否かを判定する。信頼性の高さは、例えば、予め用意したデータを用いて推論を行い、この結果からどの程度乖離しているかを算出してもよい。 When the inference result is acquired in step 6, it is next determined whether or not the inference result is highly reliable (S8). Here, the reliability of the inference result in step S6 is calculated, and it is determined whether or not this reliability is higher than the predetermined value. For example, the high reliability may be inferred using data prepared in advance, and the degree of deviation from this result may be calculated.
　ステップＳ８における判定の結果、信頼性が低い場合、またはステップＳ４における判定の結果、推論モデルがなかった場合には、学習用データを収集し、推論モデルを作成する（Ｓ１０）。この場合には、制御部１は、ステップＳ３において取得した特定情報と類似データを収集する。この場合、インターネット上に接続されている多数のＤＢ部８や関連検査機関９等内のデータベース等、種々のデータベースを検索し、上述の類似データを収集する。この場合、病名や検査項目等が分かるデータであることが望ましい。また、どのような生体情報であるかが分かる（或いは、どのような仕様、性質の測定機器、センサ等によって測定したかが分かる）時系列データであり、また想定する疾患の予兆、前兆、あるは疾患にかかったことが分かるだけの時系列パターンを読み取れるだけのデータ数や取得の時間レンジがあることが好ましい。 If the result of the determination in step S8 is low reliability, or if the result of the determination in step S4 is that there is no inference model, training data is collected and an inference model is created (S10). In this case, the control unit 1 collects the specific information and similar data acquired in step S3. In this case, various databases such as a large number of DB units 8 connected on the Internet and databases in related inspection institutions 9 and the like are searched, and the above-mentioned similar data is collected. In this case, it is desirable that the data shows the name of the disease, test items, and the like. In addition, it is time-series data that shows what kind of biological information it is (or what kind of specifications and properties it is measured by measuring devices, sensors, etc.), and also has signs and precursors of assumed diseases. It is preferable that the number of data and the acquisition time range are sufficient to read the time-series pattern that indicates that the patient has a disease.
　ステップＳ１０において、類似データを収集すると、個々の類似データについて、入力データに出力結果を関連付けて教師データを作成する。このパターンに対するアノテーション情報としては、診察されるに至った医療機関の情報（診療科やそこにある設備、備品などまで分かるようなものが含まれていればさらに好ましい）があるかないかといった情報を想定している。これらのパターンとアノテーション情報を用いて生成した教師データを集める。 When similar data is collected in step S10, teacher data is created by associating the output result with the input data for each similar data. As annotation information for this pattern, information such as whether or not there is information on the medical institution that led to the medical examination (more preferable if it includes information such as the clinical department and the equipment and fixtures there). I'm assuming. Teacher data generated using these patterns and annotation information is collected.
　教師データが作成できると、制御部１は推論依頼部１ｅを通じて学習部５に推論モデルの作成を依頼する。この教師データによって学習した推論モデルは、信頼性高く、そのデータパターンの人が健康を維持できるか、医者の診察を受けることになるか等に使用することができる。また、既存の推論モデルの信頼性が低い時に、新しい推論モデルを、教師データから作り直すようにしているので、常に、信頼性の高い推論を行うことが可能となる。学習部５が推論モデルを作成すると、制御部１に送信してくる。制御部１は推論モデルを受信すると、この推論モデルを用いて、ステップＳ４～Ｓ８を実行する。 When the teacher data can be created, the control unit 1 requests the learning unit 5 to create an inference model through the inference request unit 1e. The inference model learned from this teacher data is highly reliable and can be used to determine whether a person with that data pattern can maintain good health, see a doctor, and so on. In addition, when the reliability of the existing inference model is low, a new inference model is recreated from the teacher data, so that highly reliable inference can always be performed. When the learning unit 5 creates an inference model, it sends it to the control unit 1. When the control unit 1 receives the inference model, it executes steps S4 to S8 using this inference model.
　なお、ユーザが医療施設に診察を受けに行く時期等を推論するための推論モデルに限らず、このユーザの状態では医療施設に行くことになってしまうかもしれないが、行かずに済ますための対処方法をアドバイスするような推論モデルを生成してもよい。例えば、インターネット上のＳＮＳやブログ等に、病気にかかりそうだったが手当てしたこと等の記載があれば、これらの情報を集めてもよい。また、医療データベース等に病気にかかりそうな場合の手当て等の情報が記載されていれば、これらの情報を集めてもよい。情報を収集できれば、これらの情報を用いて学習し、推論モデルを生成する。 In addition, it is not limited to the inference model for inferring the time when the user goes to the medical facility for medical examination, etc. In this user's state, it may end up going to the medical facility, but to avoid going. You may generate an inference model that advises you on how to deal with it. For example, if there is a description in the SNS or blog on the Internet that you are likely to get sick but have been treated, you may collect this information. In addition, if the medical database or the like contains information such as treatment for a person who is likely to get sick, such information may be collected. If information can be collected, learning is performed using this information and an inference model is generated.
　図１に示す関連検査機関９として、前述のように、特定のカテゴリーに分類された健康関連情報を選択、記載することによって、ユーザの健康回復、維持に資する情報を導き出して提示する技術に基づくサービスを想定してもよい。したがって、このようなサービスを利用したい場合は、これを前提とした推論モデルを作成するようにしてもよい。この場合、そのサービスを行っている企業や組織のサービス画面のＵＲＬ情報と、そのページ、あるいはそれに続くページの特定の書き込み位置、あるいは選択位置で、記入すべき、あるいは選択すべき項目を推論して導き出すような推論モデルを作成すればよい。つまり、教師データとして特定の対象者の履歴データ（図２の時系列データ参照）と合わせて、その人（またはその人の関係者、医療従事者など）が入力した書き込み、選択情報をアノテーションすれば、推論モデルの教師データ化が可能となる。 As the related inspection organization 9 shown in FIG. 1, as described above, it is based on a technique of deriving and presenting information that contributes to the recovery and maintenance of the user's health by selecting and describing health-related information classified into a specific category. You may assume a service. Therefore, if you want to use such a service, you may create an inference model based on this. In this case, infer the URL information of the service screen of the company or organization that provides the service, and the item to be entered or selected at the specific writing position or selection position of the page or the page following it. You can create an inference model that can be derived from the above. That is, the writing and selection information input by the person (or the person concerned, the medical staff, etc.) should be annotated together with the historical data of the specific target person (see the time series data in FIG. 2) as the teacher data. For example, it is possible to convert the inference model into teacher data.
　近年は、医師、医療従事者も上述したようなサービスを利用して診断の参考にする場合があるので、特定の患者の診察時等に医師が行った、診察・診断に際しての医師の利用するＰＣ等の情報端末への手動入力結果、または音声入力結果を利用してもよい。その患者の履歴データ（図２の時系列データ参照）を取得し、この履歴データに上述の入力結果を、どの項目がどのような入力情報に対応するかを識別可能にアノテーション情報として付与していく。患者の履歴データは、必ずしも一つの項目の履歴データである必要はなく、時系列的に生活習慣の情報や通院、服薬の情報などを記録したものなどを入れ込んで利用してもよい。この作業によって、類似したパターンで生活する人のデータを参照することができ、推論の精度が高まる。こうして得られた教師データを使って、信頼性の高い結果が出るように学習を行って、新しい推論モデルを作成する。 In recent years, doctors and medical professionals may also use the services described above to refer to the diagnosis, so it is used by doctors at the time of medical examination / diagnosis performed by doctors at the time of medical examination of a specific patient. The manual input result to an information terminal such as a PC or the voice input result may be used. The patient's history data (see the time-series data in FIG. 2) is acquired, and the above-mentioned input result is added to this history data as annotation information so that which item corresponds to what kind of input information can be identified. I will go. The patient history data does not necessarily have to be the history data of one item, and may be used by incorporating information on lifestyle habits, hospital visits, medication information, etc. in chronological order. By this work, it is possible to refer to the data of people who live in a similar pattern, and the accuracy of inference is improved. Using the teacher data obtained in this way, we train to obtain reliable results and create a new inference model.
　つまり、このような工夫によって、特定の仕様の検査機器を用いて時系列的に検査データを取得し、該検査データの変化パターン情報を取得し、この変化パターン情報に対し、特定の健康アドバイスサービスの入力情報をアノテーションした結果を教師データとして学習させた推論モデルに対し、特定の人物の特定の仕様の検査機器を用いて時系列的に得た検査データを入力して推論した結果を、特定の人物に伝達する情報伝達方法が提供可能となる。 That is, by such a device, inspection data is acquired in time series using an inspection device having a specific specification, change pattern information of the inspection data is acquired, and a specific health advice service is provided for this change pattern information. For the inference model in which the result of annotating the input information of is trained as teacher data, the result of inferring by inputting the inspection data obtained in time series using the inspection equipment of the specific specifications of a specific person is specified. It becomes possible to provide a method of transmitting information to the person in question.
　言い換えると、この情報伝達方法は、まず、特定の健康対応イベントの入力に対応し、健康対応イベントに先立って特定の仕様の検査機器を用いて、特定の健康対応イベントに対応して、時系列的に検査データを取得し、この時系列的な検査データを記録する。そして、記録した検査データの変化パターン情報に、健康対応イベントの行われた時の設備、備品、環境の情報をアノテーションした結果を教師データとして作成する。この作成された教師データを用いて、学習を行い、推論モデルを生成する。特定の人物について、特定の仕様の検査機器を用いて時系列的に検査データを取得し、取得した検査データを推論モデルに入力して、設備、備品、環境の情報を推論する。この推論結果から特定の人物用にカスタマイズした健康対応イベントを、特定の人物に伝達する。ここで、健康対応イベントは、健康にかかわるようなイベントであり、例えば、医療施設において医師の診察を受ける等のイベントである。 In other words, this information transmission method first responds to the input of a specific health response event, and prior to the health response event, uses a specific specification inspection device to respond to the specific health response event in chronological order. The inspection data is acquired and the time-series inspection data is recorded. Then, the result of annotating the change pattern information of the recorded test data with the information of the equipment, equipment, and environment at the time of the health response event is created as teacher data. Using this created teacher data, learning is performed and an inference model is generated. For a specific person, inspection data is acquired in chronological order using an inspection device with specific specifications, and the acquired inspection data is input to an inference model to infer information on equipment, equipment, and the environment. From this inference result, a health response event customized for a specific person is transmitted to a specific person. Here, the health response event is an event related to health, for example, an event such as receiving a medical examination by a doctor at a medical facility.
　なお、推論モデルを生成するためには、適正規模の教師データからなる学習用母集団を作成しなければならない。このために、多数の個人の情報を収集して学習用母集団を作成する。この場合、検査機器の種類や、また検査を行った期間等の情報を含めて収集し、これらの情報毎に学習用母集団を作成するとよい。 In addition, in order to generate an inference model, it is necessary to create a learning population consisting of teacher data of an appropriate scale. For this purpose, a large number of individual information is collected to create a learning population. In this case, it is advisable to collect information such as the type of inspection equipment and the period during which the inspection was performed, and create a learning population for each of these pieces of information.
　ステップＳ８における判定の結果、信頼性が高い場合には、次に、推論結果に基づいて推奨施設を表示する（Ｓ１３ａ）。ここでは、ステップＳ６における推論結果に基づいて、推奨施設を端末４に表示させる。ステップＳ６において、必要な検査項目を推論した場合には、この検査を実施できる医療施設や検査機関をＤＢ８内の施設別保有機器一覧８ａを検索して、検索結果を表示すればよい。また、ステップＳ６において、検査可能な医療施設や検査機関を推論できる場合には、推論結果に基づいて推奨施設を表示すればよい。 If the result of the determination in step S8 is high reliability, then the recommended facility is displayed based on the inference result (S13a). Here, the recommended facility is displayed on the terminal 4 based on the inference result in step S6. When the necessary inspection items are inferred in step S6, the medical facilities and inspection institutions that can carry out this inspection may be searched for the facility-specific possessed equipment list 8a in the DB8, and the search results may be displayed. Further, in step S6, if it is possible to infer a medical facility or an institution that can be inspected, the recommended facility may be displayed based on the inference result.
　ステップＳ１３ａでは、時系列的に得られた個人ごとの生体情報検査データを利用して、推奨施設等のアドバイスを取得している。さらに、その生体情報検査データの変化パターン情報に対し、改善効果のあった医療機関等の特徴情報をアノテーションした結果、もしくは、特定の健康アドバイスサービスの入力情報をアノテーションした結果を教師データとして学習させた推論モデルを生成し、この推論モデルを利用して検査データの改善時のきっかけ情報を得る方法について説明している。 In step S13a, advice on recommended facilities, etc. is obtained using the biometric information test data for each individual obtained in chronological order. Furthermore, the result of annotating the characteristic information of medical institutions, etc. that had an improvement effect on the change pattern information of the biometric information test data, or the result of annotating the input information of a specific health advice service is learned as teacher data. The method of generating the inference model and using this inference model to obtain the trigger information when improving the inspection data is explained.
　また、ここで得られる検査データの改善時のきっかけ情報は、検査や処置のレベルにまで還元した情報であることから、情報を伝達された人（情報を受ける人）の、生活パターンや、生活する地域、地方においても活用可能な情報が伝達されるように変換してアドバイスを提示する工夫が望まれる。情報を受ける人の生活に照らし合わせて、条件を満たすもの探すことで、他の地方においても活用可能な情報として取得可能となる。情報を受け取る人の生活圏、もしくはその隣接エリアを条件に加えて、抽出条件を満たす施設等を検索して該当するものを表示などで伝達すればよい。 In addition, since the information that triggers the improvement of the test data obtained here is information that has been reduced to the level of the test or treatment, the life pattern and life of the person who transmitted the information (the person who receives the information). It is desirable to devise ways to convert and present advice so that information that can be used is transmitted even in the regions and regions where it is used. By searching for information that meets the conditions in light of the lives of the people who receive the information, it will be possible to obtain information that can be used in other regions as well. In addition to the living area of the person who receives the information or the area adjacent to it, it is sufficient to search for facilities that satisfy the extraction conditions and convey the corresponding items by display or the like.
　また、将来的にかかりつけ医など、近所の診療機関で、ワンストップサービスで、様々な医療サービスを受けられるような社会となる可能性がある。この場合は、専門の医師の遠隔の指示などを受けて診断・治療の計画などが立てられる。したがって、このアドバイスは、かかりつけ医に、風邪などの比較的軽い病状で行った場合等に、そこの医療従事者や端末等で発せられ、伝えられてもよい。この場合、風邪で診察を受けつつも、これまでの生体データの変化パターンなどで、将来のリスクが分かるので、リスクが改善されるような遠方の医師のアドバイスを取り次いでもらうようなサービスにしてもよい。 In the future, there is a possibility that a society will be created in which various medical services can be received with a one-stop service at a nearby medical institution such as a family doctor. In this case, a diagnosis / treatment plan can be made by receiving remote instructions from a specialist doctor. Therefore, this advice may be given and conveyed to the family doctor by the medical staff, the terminal, or the like when the patient has a relatively mild medical condition such as a cold. In this case, even if you are examined for a cold, you can know the future risk from the change pattern of biological data so far, so make it a service that asks for advice from a distant doctor who can improve the risk. May be good.
　簡単な生活改善等、あるいは検査であれば、例えば、その人の近所のコンビニエンスストアなどで行うサービスの一環としての処置も出てくるものと考えられる。少なくとも、自覚症状のない未病状態であれば、このようなサービスによるアドバイスを受けて、食事内容の調整など生活習慣の改善で状況が改善可能になる。また、必要な追加検査などがある場合は、コンビニエンスストア等において予約の手続きや支払いが出来るようにすれば良い。これらのサービスによって、生体データで表される健康状況が改善されれば、それもまた教師データ化を簡単にすることができ、同様の未病の人を簡単なアドバイスで病気の発症を防ぐことが可能となる。 If it is a simple life improvement or inspection, for example, it is thought that treatment as part of the service provided at a convenience store in the neighborhood of the person will come out. At the very least, if you are in a non-illness state without any subjective symptoms, you can improve the situation by receiving advice from such services and improving your lifestyle such as adjusting your diet. In addition, if there is a necessary additional inspection, it is sufficient to make reservation procedures and payment at a convenience store or the like. If these services improve the health status represented by biometric data, it can also facilitate teacher data conversion and prevent the onset of illness with simple advice for similar non-illness people. Is possible.
　ステップＳ１３ａにおいて、推奨施設の表示を行うと、ステップＳ１に戻る。 When the recommended facility is displayed in step S13a, the process returns to step S1.
　このように、図７に示した検査結果送信のフローにおいては、特定の入力データに対して特定の推論モデルの推論結果の信頼性が低い場合に（Ｓ８参照）、新たに特定の入力データと類似の入力データを収集して教師データを作成し、この教師データを用いて学習を行い（Ｓ１０参照）、この学習によって生成された新たな推論モデルを用いて推論した結果を採用している（Ｓ１０→Ｓ４～Ｓ８Ｙｅｓを参照）。 As described above, in the inspection result transmission flow shown in FIG. 7, when the inference result of the specific inference model is unreliable with respect to the specific input data (see S8), the inference result is newly added to the specific input data. Similar input data is collected to create teacher data, training is performed using this teacher data (see S10), and the result of reasoning using the new reasoning model generated by this training is adopted (see S10). See S10 → S4 to S8Yes).
　次に、推論モデル作成の動作について説明する。本実施形態においては、種々のステップ（例えば、図４のＳ９、図５のＳ３５、図７のＳ６）において、推論モデルを用いて推論を行っている。推論モデルは、生体情報の時系列変化のパターンを入力し、将来の健康リスクやその改善策を推論し、あるいはそれがどれくらい将来であるかの推論を可能とすることもできる。図８Ａ、図８Ｂを用いて、このような推論モデルの作成について、具体的に説明する。 Next, the operation of creating an inference model will be explained. In the present embodiment, inference is performed using an inference model in various steps (for example, S9 in FIG. 4, S35 in FIG. 5, and S6 in FIG. 7). The inference model can also input patterns of time-series changes in biometric information to infer future health risks and their remedies, or to infer how far in the future it is. The creation of such an inference model will be specifically described with reference to FIGS. 8A and 8B.
　図８Ａは、健康状態の時間的変化を示し、横軸は時間Ｔの変化を示し、縦軸は生体データＤを示す。この生体データＤは値が大きくなると健康状態が悪化し、生体データＤの値が小さくなると健康状態が改善する。図８Ａの黒丸が連なった時系列データＤｔｓ１１、Ｄｔｓ１２は、何かのきっかけ（特定イベント）で健康状態が悪化した場合を示し、白丸が連なった時系列データＤｔｓ２１、Ｄｔｓ２２は何かのきっかけ（特定イベント）で健康状態が改善した場合を示す。すなわち、健康状態は何かのきっかけ（特定イベント；タイミングをきっかけ情報Ｉｎｆで示す）によって、改善しまたは悪化し、健康状態が生体データＤに現れて、パターンのトレンドが変化する。なお、図８Ａにおいて、きっかけ情報Ｉｎｆと平行な点線で示した縦線は、他のイベントの例である。 FIG. 8A shows the temporal change of the health state, the horizontal axis shows the change of time T, and the vertical axis shows the biological data D. When the value of the biological data D becomes large, the health condition deteriorates, and when the value of the biological data D becomes small, the health condition improves. The time-series data Dts11 and Dts12 in which the black circles in FIG. 8A are connected indicate the case where the health condition deteriorates due to some trigger (specific event), and the time-series data Dts21 and Dts22 in which the white circles are connected are the triggers (specification). The event) shows when the health condition is improved. That is, the health condition is improved or deteriorated by some trigger (specific event; timing is indicated by the trigger information Inf), the health condition appears in the biological data D, and the trend of the pattern changes. In FIG. 8A, the vertical line shown by the dotted line parallel to the trigger information Inf is an example of another event.
　前述したように、仮に生体データＤの上昇が健康状態からの乖離を表し、生体データＤの下降は回復、改善を表すとする。例えば、何か悪いものを食べて（これがイベント）発熱したとか、診察を受けて薬を服用して（これがイベント）回復したとかが身近にありうるが、多くの場合、何が引き金になっているか分からなかったり、どのように回復したりしたかを数値的に意識して把握している患者は少ない。 As described above, it is assumed that an increase in biological data D indicates a deviation from the health state, and a decrease in biological data D indicates recovery or improvement. For example, eating something bad (this is an event) and having a fever, or having a medical examination and taking medicine (this is an event) and recovering can be familiar, but in many cases what is the trigger. There are few patients who do not know or are aware of how they have recovered numerically.
　もちろん、図８Ａに示す生体データＤが、上昇または下降のどちらの方向に変化するかと、健康的かどうかになるは、扱う数値やその程度によって異なる。例えば、図８Ａでは、以下のようなことが表現できる。時系列データＤｔｓ１１、Ｄｔｓ１２のパターンが（微小な数値変化を伴いつつも）数値低下のトレンド１（Ｄｔｓ１１）である場合、生活習慣などで体調が変化していると考えられる。それに加えてさらなる体調悪化イベント（例えば睡眠不足や暴飲暴食や、風邪などの影響）（Ｉｎｆのタイミング）によって、さらにその傾向がひどくなってトレンド２（Ｄｔｓ１２）になった様子を表現しているものとも言える。また、時系列データＤｔｓ２１、Ｄｔｓ２２は別の数値であり、健康状態の変化（例えば老化による筋力とか骨密度や視力）で低下していたものが（Ｄｔｓ２１）、病院に行って治療を受けるといったイベント（Ｉｎｆのタイミング）によって、それが回復した様子を示す例として捉えても良い。 Of course, whether the biological data D shown in FIG. 8A changes in the ascending or descending direction and whether it is healthy or not depends on the numerical value to be handled and the degree thereof. For example, in FIG. 8A, the following can be expressed. When the pattern of the time-series data Dts11 and Dts12 is the trend 1 (Dts11) of the numerical decrease (although accompanied by a slight numerical change), it is considered that the physical condition has changed due to lifestyle habits and the like. In addition to that, it expresses how the tendency became worse and became trend 2 (Dts12) due to further deterioration events (for example, the effects of lack of sleep, overdrinking and eating, colds, etc.) (timing of Inf). It can be said that. In addition, the time-series data Dts21 and Dts22 are different numerical values, and those that have decreased due to changes in health status (for example, muscle strength, bone density, and visual acuity due to aging) (Dts21) are events such as going to a hospital and receiving treatment. (Inf timing) may be taken as an example showing how it has recovered.
　そこで、生体データＤをモニタリングしていて、かつ、その人の健康に反映されるイベントを、この時系列生体データパターン判定と同様の時間座標に関連付けて記録できるデータベースがあれば、図８Ｂに示すようなフローチャートを実行することによって、何がどのように健康を左右するかのイベントを特定することが可能となる。図８Ｂに示すフローは、制御部１内のＣＰＵ等によって、制御部１内の各部を制御することによって実行するが、制御部１に限らず、他のサーバ、機関等によって実行してもよい。 Therefore, if there is a database that monitors the biometric data D and can record the events reflected in the health of the person in association with the time coordinates similar to the time series biometric data pattern determination, it is shown in FIG. 8B. By executing such a flowchart, it is possible to identify an event of what affects health and how. The flow shown in FIG. 8B is executed by controlling each unit in the control unit 1 by a CPU or the like in the control unit 1, but is not limited to the control unit 1 and may be executed by another server, engine, or the like. ..
　図８Ｂに示す推論モデル作成のフローが開始すると、まず、時系列データ取得期間に対応するイベント情報を取得する（Ｓ６１）。時系列データ取得期間は、対象となる疾病や健康状態等によって変化するが、ここでは、一般的にイベントが健康に影響する期間を考慮して決めればよい。また、イベント情報は、ユーザの端末４に記録されている予定や、ＳＮＳ等にアップロードされた予定等、種々のソースから取得する。このステップでは、特定の仕様の検査機器を用いて特定の時系列的に検査データを取得している。なお、イベントは、前述したように、対象者が病院等に行って診察を受けたことや、薬局で薬を購入して服薬したことや、寒くて人混みの場所に行ったこと等、対象者の健康状態に影響を与える可能性のある行為や行動等である。 When the flow of creating the inference model shown in FIG. 8B starts, first, the event information corresponding to the time series data acquisition period is acquired (S61). The time-series data acquisition period varies depending on the target disease, health condition, etc., but here, it may be determined in consideration of the period in which the event generally affects the health. Further, the event information is acquired from various sources such as a schedule recorded on the user's terminal 4 and a schedule uploaded to SNS or the like. In this step, inspection data is acquired in a specific time series using an inspection device having specific specifications. As mentioned above, the event was attended by the subject, such as going to a hospital for medical examination, purchasing medicine at a pharmacy and taking it, or going to a cold and crowded place. Actions and behaviors that may affect your health.
　続いて、特定イベント前後のパターンのトレンド変化を比較し、差異があるか否かを判定する（Ｓ６３）。ステップＳ６１において特定イベント情報が取得された場合に、特定イベントの前後において、時系列データ（生体データ）のトレンドに変化があるか否かを判定する。すなわち、検査データの変化パターン情報を取得しており、また特定イベント前の時系列データのパターンと特定イベント後の時系列データのトレンド（傾向）が同じであるか否かを判定する。 Subsequently, the trend change of the pattern before and after the specific event is compared, and it is determined whether or not there is a difference (S63). When the specific event information is acquired in step S61, it is determined whether or not there is a change in the trend of the time series data (biological data) before and after the specific event. That is, the change pattern information of the inspection data is acquired, and it is determined whether or not the pattern of the time series data before the specific event and the trend (trend) of the time series data after the specific event are the same.
　ステップＳ６３における判定の結果、トレンド変化に差異がない場合には、別イベントを用いて判定する（Ｓ７５）。この場合は、ステップＳ６３におけるイベント前後でトレンド変化に差異がないことから、イベント前後で健康状態に変化ないといえる。そこで、ステップＳ６３に戻り、トレンド変化のあるイベントを探す。 If there is no difference in the trend change as a result of the determination in step S63, the determination is made using another event (S75). In this case, since there is no difference in the trend change before and after the event in step S63, it can be said that the health state does not change before and after the event. Therefore, the process returns to step S63, and an event with a trend change is searched for.
　一方、ステップＳ６３における判定の結果、特定イベントの前後で、トレンドに変化があった場合には、きっかけ情報（図８ＡのＩｎｆ）前後におけるデータ変化のトレンドを分類する（Ｓ６５）。きっかけ情報前後にトレンド変化に差異がない時とある時で、イベントが健康に影響したかどうかが分かる。ただし、それは良い影響のこともあれば悪い影響の場合もあるので、このステップＳ６５において、良いトレンドか悪いトレンドかを判定し、この判定結果に基づいて、データ変化パターンのトレンドを分類する。 On the other hand, if there is a change in the trend before and after the specific event as a result of the determination in step S63, the trend of the data change before and after the trigger information (Inf in FIG. 8A) is classified (S65). Trigger information You can see whether the event affected your health when there is no difference in the trend change before and after. However, since it may have a good effect or a bad effect, in this step S65, it is determined whether the trend is a good trend or a bad trend, and the trend of the data change pattern is classified based on the determination result.
　続いて、改善イベントと悪化イベントに分類し、そのイベントの環境、構成要素をアノテーション情報とする（Ｓ６７）。ここでは、ステップＳ６５における分類結果を、良くなったイベントか悪化したイベントかに分類し、さらにそのイベントの環境や構成要素等をアノテーション情報とする。環境や構成要素は、例えば、単に「東京駅に行った」というイベントを、寒い日に、人混みの場所に行ったというように、イベントを分解したものをいう。また、単に、「病院に行って診察を受けた」というイベントを、例えば、「レントゲン撮影を行った」、「〇〇を注射した」、「△△を服用した」等、イベントで行ったことを分解したものをいう。すなわち、ステップＳ６７においては、イベント情報を構成毎に分解して構成要素を抽出している。構成要素等を抽出し、教師データを作成し、学習することによって、トレンド変化に影響したイベント情報を抽出した構成要素に基づいてカスタマイズした情報提供が可能となる。 Subsequently, it is classified into an improvement event and a deterioration event, and the environment and components of the event are used as annotation information (S67). Here, the classification result in step S65 is classified into an improved event or a worsened event, and the environment, components, and the like of the event are used as annotation information. The environment and components are the decomposition of an event, such as simply going to a crowded place on a cold day, instead of simply going to Tokyo Station. In addition, the event "I went to the hospital and had a medical examination" was simply held at an event such as "I took an X-ray", "I injected XX", "I took △△", etc. It is a disassembled version of. That is, in step S67, the event information is decomposed for each configuration and the components are extracted. By extracting the components and the like, creating teacher data, and learning, it is possible to provide customized information based on the extracted components that have affected the trend change.
　続いて、特定イベント前、または特定イベント後のトレンド変化を示すパターンを教師データ化する（Ｓ６９）。ここでは、ステップＳ６５における分類に基づいて、特定イベント前のトレンド変化を示す時系列情報パターンに対して、その良いイベントと悪いイベントをアノテーションして教師データ化する。 Subsequently, the pattern showing the trend change before or after the specific event is converted into teacher data (S69). Here, based on the classification in step S65, the good event and the bad event are annotated with respect to the time-series information pattern indicating the trend change before the specific event and converted into teacher data.
　ステップＳ６９において、教師データが作成されると、次に、学習・推論モデルの作成を行う（Ｓ７１）。ここでは、ステップＳ６９において作成された教師データを用いて学習し推論モデルを生成する。すなわち、トレンドが変化する場合には、きっかけとなるイベントがあることから、時系列データを入力した際に、特定イベント（影響イベント）を出力することができる推論モデルを生成する。また、ステップＳ６９において、特定イベント前のパターンを用いて教師データを生成すれば、その特定イベントがあると（こんな風に）悪化するという推論を行うことができる推論モデルを生成できる。一方、特定イベント後のパターンを用いて教師データ生成すれば、その特定イベントがあれば、（こんな風に）改善するという推論を行うことができる推論モデルを生成できる。 When the teacher data is created in step S69, then the learning / inference model is created (S71). Here, learning is performed using the teacher data created in step S69 to generate an inference model. That is, when the trend changes, there is an event that triggers it, so when time series data is input, an inference model that can output a specific event (impact event) is generated. Further, in step S69, if the teacher data is generated using the pattern before the specific event, it is possible to generate an inference model capable of inferring that the presence of the specific event makes it worse (in this way). On the other hand, if teacher data is generated using the pattern after a specific event, it is possible to generate an inference model that can make an inference that if there is the specific event, it will be improved (in this way).
　つまり、このステップＳ７１において、変化パターン情報に対し、改善効果のあった医療機関等の特徴情報や、特定期間内で変換パターンが変化した時点に対応する時点における影響イベント情報を検出し、検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得している。この推論モデルは、どのようなイベントによって健康が改善、または悪化するかを判定できる。 That is, in this step S71, with respect to the change pattern information, the characteristic information of the medical institution or the like having the improvement effect and the influence event information at the time corresponding to the time when the conversion pattern changes within a specific period are detected and detected. We have acquired an inference model in which the result of annotating the impact event information is trained as teacher data. This inference model can determine what events improve or worsen health.
　推論モデルを生成すると、特定タイミングで更新を行う（Ｓ７３）。生体データやイベント情報は、刻々と蓄積されていくので、定期的に、あるいは時候やニュース等、何かのきっかけで更新を行う。ステップＳ７１において生成した推論モデルに、特定データパターンを入力すれば、将来、このイベントで改善できる、あるいは悪化するから要注意という意識付けが出来るアドバイスが可能となる。季節の変化や、感染傾向のある疾病などについては混雑地域への移動などの情報もイベントと採用した方が良い。これは、時系列データの時間軸に対応させて、ステップＳ７３において追記できるようにしてもよい。また、イベント後のデータ変化で、何が悪かったかを判定する推論モデルも作成できる。 When an inference model is generated, it is updated at a specific timing (S73). Biological data and event information are accumulated from moment to moment, so they are updated regularly or at the time of the weather, news, etc. If a specific data pattern is input to the inference model generated in step S71, it is possible to give advice that can be improved or deteriorated in this event in the future, so that it is necessary to be aware of it. For seasonal changes and diseases that are prone to infection, it is better to use information such as moving to congested areas as an event. This may be added in step S73 so as to correspond to the time axis of the time series data. You can also create an inference model that determines what went wrong with data changes after the event.
　このように、推論モデル作成のフローは、健康状態を表す生体情報の時系列データの変化パターンに影響するイベントを判定するステップ（Ｓ６３）と、そのイベントが改善効果があるかを判定するステップ（Ｓ６７）と、そのイベントの時間の前の時系列パターンに従って、変化パターンの改善効果のあるイベントを判定するステップ（Ｓ６９）とを有する情報提供方法を示している。すなわち、イベント後のパターンに基づいて、健康に影響したイベントを発見することができる。 In this way, the flow of creating the inference model is a step of determining an event that affects the change pattern of the time-series data of the biological information representing the health state (S63) and a step of determining whether the event has an improvement effect (S63). An information providing method including S67) and a step (S69) of determining an event having an effect of improving a change pattern according to a time-series pattern before the time of the event is shown. That is, it is possible to discover events that affect health based on the post-event pattern.
　これを言い換えると、本実施形態における記録システムは、特定の仕様の（生体データ）検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得して作成したデータの変化パターン情報に対し、この特定の期間内におけるデータ取得元に関する複数のイベント情報を、それぞれ何時起こったかのタイミングまで関連付けて記録することができる。この記録システムは、データやイベントを記録するデータベース等を含んでいてもよい。また、このデータベースは連携した複数のコンピュータであってもよい。この記録システムは、トレンド変化に影響したイベントが抽出できるはずなので、この影響があったイベントの情報提供を可能とする回路やプログラムやシステムとすることによって、他の人に有益な情報を提供しうる。 In other words, the recording system in the present embodiment acquires a plurality of inspection data in chronological order over a specific period using a (biological data) inspection device having a specific specification, and creates a data change pattern. With respect to the information, a plurality of event information regarding the data acquisition source within this specific period can be recorded in association with each other up to the timing of when it occurred. This recording system may include a database for recording data and events. Further, this database may be a plurality of linked computers. Since this recording system should be able to extract the events that affected the trend change, it provides useful information to other people by making it a circuit, program, or system that can provide information on the events that have affected this trend. sell.
　イベントのあった環境や状況やその他得られる構成要素情報に基づいて、イベント情報を細分化すれば、情報提供する人の立場に置き換えたカスタマイズ情報として提供することができる。例えば、感染症発生が報告された地域の人混みに行った、といった要素に分解すれば、フェリー乗り場に行ったという実際の具体的行動情報より一般化でき、今日は混雑した場所には行くな、といった情報に変換して伝えられる。また、検査データ変化パターンと、こうした影響イベントのアノテーションによって得られた多数の人のケースの教師データを集め、学習し推論モデルを作成し、推論モデルに変化パターンを入力し、出力を影響イベント要素とするようにしてもよい。 If the event information is subdivided based on the environment and situation where the event occurred and other component information obtained, it can be provided as customized information replaced by the position of the person who provides the information. For example, if you break it down into factors such as going to a crowd in the area where the outbreak of an infectious disease was reported, it can be generalized from the actual specific behavioral information that you went to the ferry landing, so don't go to a crowded place today. It is converted into information such as. In addition, the inspection data change pattern and the teacher data of many cases obtained by annotation of these influence events are collected, learned and an inference model is created, the change pattern is input to the inference model, and the output is the influence event element. You may try to.
　なお、ここでは、イベントを順次判定して、パターンのトレンド変化に対応するものを探している。これは、上述の考え方の思想を表したものであって、それ以外の方法をとってもよい。例えば、トレンド変化を検出してから、そのタイミングに対応するイベント（あるいはその構成要素、要因、環境等）を検索してもよい。このイベントとして医療行為等を想定すれば、健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報が、この構成要素、要因、環境等に相当する。 Here, we are looking for something that corresponds to the trend change of the pattern by sequentially judging the events. This expresses the idea of the above-mentioned idea, and other methods may be taken. For example, after detecting a trend change, an event (or its component, factor, environment, etc.) corresponding to the timing may be searched. Assuming medical practice as this event, the equipment and / or equipment and / or environmental information at the time of the health response event corresponds to this component, factor, environment, and the like.
　以上、説明したように、本発明の一実施形態においては、特定の仕様の検査機器を用いて特定の時系列的に検査データを取得し（例えば、図８（ｂ）のＳ６１参照）、この検査データの変化パターン情報を取得し（例えば、図８（ｂ）のＳ６３参照）、変化パターン情報に対し、改善効果のあった医療機関等の特徴情報や、特定期間内で変換パターンが変化した時点に対応する時点における環境イベント情報を検出し、検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得している（例えば、図８（ｂ）のＳ６５～Ｓ７１参照）。そして、特定の人物の特定の仕様の検査機器を用いて、特定期間の幅と類似の期間において時系列的に得た検査データを、推論モデルに入力して影響イベントを推論し、該推論の結果を伝達している（例えば、図４のＳ９、図５のＳ３５、図７のＳ６において、上述の推論モデルを用いて推論することができる）。 As described above, in one embodiment of the present invention, inspection data is acquired in a specific time series using an inspection device having specific specifications (see, for example, S61 in FIG. 8B). The change pattern information of the test data was acquired (for example, see S63 in FIG. 8 (b)), and the characteristic information of the medical institution etc. that had an improvement effect on the change pattern information and the conversion pattern changed within a specific period. We have acquired an inference model in which the environmental event information at the time corresponding to the time point is detected and the result of annotating the detected influence event information is trained as teacher data (see, for example, S65 to S71 in FIG. 8B). ). Then, using an inspection device of a specific person and a specific specification, the inspection data obtained in time series in a period similar to the width of the specific period is input to the inference model to infer the influence event, and the inference is performed. The results are transmitted (for example, in S9 in FIG. 4, S35 in FIG. 5, and S6 in FIG. 7, it can be inferred using the above-mentioned inference model).
　また、本発明の一実施形態においては、特定の健康対応イベントの入力に対応し、健康対応イベントに先立って特定の仕様の検査機器を用いて時系列的に検査データを取得し記録されていた、該検査データの変化パターン情報に対し、健康対応イベントの行われた時の設備や、備品や、環境の情報をアノテーションした結果を教師データとして学習させた推論モデルを取得している（例えば、図８（ｂ）のＳ７１参照）。そして、特定の人物の特定の仕様の検査機器を用いて時系列的に得た検査データを、上述の推論モデルに入力して推論し、該推論結果を特定の人物に伝達している（例えば、図４のＳ９、図５のＳ３５、図７のＳ６において、上述の推論モデルを用いて推論することができる）。 Further, in one embodiment of the present invention, in response to the input of a specific health response event, inspection data is acquired and recorded in time series using an inspection device having a specific specification prior to the health response event. , An inference model is acquired in which the result of annotating the equipment, equipment, and environment information at the time of the health response event is learned as teacher data for the change pattern information of the test data (for example,). See S71 in FIG. 8 (b)). Then, the inspection data obtained in time series using the inspection device of the specific specification of the specific person is input to the above-mentioned inference model and inferred, and the inference result is transmitted to the specific person (for example). , S9 in FIG. 4, S35 in FIG. 5, and S6 in FIG. 7 can be inferred using the above-mentioned inference model).
　また、本発明の一実施形態においては、特定の仕様の検査機器を用いて時系列的に検査データを取得し、該検査データの変化パターン情報を取得し（例えば、図４のＳ１参照）、変化パターン情報に対し、服薬や生活改善の開始など特定の改善タイミング情報を同様の時間情報として含むことによって、当該改善情報以前の変化パターンを比較可能にした情報を有するデータベースを作成する。このデータベースを用いて、特定の人物の変化パターンを他の人の改善情報以前の変化パターンと比較して、改善情報を検索し、検索された改善情報を特定の人物に伝達している（例えば、図４のＳ１３において、データベースの検索結果を用いて情報を伝達することができる）。 Further, in one embodiment of the present invention, inspection data is acquired in time series using an inspection device having specific specifications, and change pattern information of the inspection data is acquired (see, for example, S1 in FIG. 4). By including specific improvement timing information such as taking medication and starting life improvement as the same time information for the change pattern information, a database having information that makes it possible to compare the change patterns before the improvement information is created. Using this database, the change pattern of a specific person is compared with the change pattern of another person before the improvement information, the improvement information is searched, and the searched improvement information is transmitted to a specific person (for example). , In S13 of FIG. 4, information can be transmitted using the search results of the database).
　また、本発明の一実施形態においては、特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得して作成した検査データの変化パターン情報と、特定の期間内における検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録可能なデータベースを作成し、データベースを用いて、変化パターンのトレンド変化とイベントのタイミングの関係に基づいて、イベント情報のうち、トレンド変化に影響したイベントの情報提供を可能としている（例えば、図８Ａおよび図８Ｂ参照）。トレンド変化に影響したイベントの情報提供に当たって、データベースを検索してもよく、また推論モデルを生成し、この推論モデルを用いて推論してもよい。なお、本発明の一実施形態においては、影響イベントや健康対応イベント等、各種のイベントについて説明しているが、これらも広義のイベントに含まれる。 Further, in one embodiment of the present invention, change pattern information of inspection data created by acquiring a plurality of inspection data in a time series using an inspection device having a specific specification and a specific inspection data. Create a database that can be recorded by associating each timing of multiple event information related to the acquisition source of inspection data within the period, and use the database based on the relationship between the trend change of the change pattern and the event timing. Among the event information, it is possible to provide information on events that have influenced the trend change (see, for example, FIGS. 8A and 8B). In providing information on events that have influenced the trend change, a database may be searched, or an inference model may be generated and inferred using this inference model. In one embodiment of the present invention, various events such as impact events and health response events are described, but these are also included in the broad sense of the event.
　また、本発明の一実施形態においては、対象者の検査データを取得し（例えば、図７のＳ１参照）、検査データが特定の入力データであるか否かを判定し（例えば、図７のＳ３参照）、この判定の結果、検査データが特定の入力データであった場合に、特定の推論モデルを用いて推論を行い（例えば、図７のＳ４Ｙｅｓ→Ｓ６参照）、特定の入力データに対して特定の推論モデルの推論結果の信頼性が低くかった場合に、新たに特定の入力データと類似の入力データを収集して教師データとして、推論モデルを生成するための学習を行い（例えば、図７のＳ８Ｎｏ→Ｓ１０参照）、学習によって生成された新たな推論モデルに従って特定データを推論した結果を採用し、この結果を伝達する（例えば、図７のＳ１０→Ｓ４～Ｓ８Ｙｅｓ→Ｓ１３ａ参照）。このため、推論モデルを用いて検査データを推論して情報を提供する場合に、推論結果の信頼性が低かった場合に、被検者以外の者の検査データを収集し、新たに推論モデル生成し、精度の高い推論を行うことができる。 Further, in one embodiment of the present invention, the inspection data of the subject is acquired (for example, see S1 in FIG. 7), and it is determined whether or not the inspection data is specific input data (for example, in FIG. 7). (See S3), and if the inspection data is specific input data as a result of this determination, inference is performed using a specific inference model (see, for example, S4Yes → S6 in FIG. 7), and the specific input data is subjected to inference. When the reliability of the inference result of a specific inference model is low, training is performed to generate an inference model by newly collecting input data similar to the specific input data and using it as teacher data (for example). (See S8No → S10 in FIG. 7), the result of inferring specific data according to a new inference model generated by learning is adopted, and this result is transmitted (see, for example, S10 → S4 to S8Yes → S13a in FIG. 7). Therefore, when inferring test data using an inference model and providing information, if the reliability of the inference result is low, the test data of a person other than the subject is collected and a new inference model is generated. However, it is possible to make highly accurate inferences.
　また、本発明の一実施形態においては、対象者の検査データを入力し（例えば、図１の情報判定機器２、図４のＳ１、図５のＳ２１等参照）、対象者のプロフィール情報と、検査・医療機関ごとの保有機器情報を取得し（例えば、図１のＤＢ部８、図４のＳ７等参照）、対象者の検査データと、プロフィール情報と、検査・医療機関ごとの保有機器情報に従って、対象者に伝達する伝達情報を決定している（例えば、図１の情報提供部１ｃ、図４のＳ１１等参照）。このため、検査・医療機関の保有機器を考慮して、対象者に情報を伝達することができる。また、本実施形態においては、日常的に無意識に得られた健康関連情報をもとに、有効な施設の紹介を可能とし、何度も施設に足を運ぶ手間を省くことができる。 Further, in one embodiment of the present invention, the inspection data of the subject is input (for example, refer to the information determination device 2 in FIG. 1, S1 in FIG. 4, S21 in FIG. 5 and the like), and the profile information of the subject and the profile information of the subject. Acquire the possessed device information for each test / medical institution (see, for example, DB section 8 in FIG. 1, S7 in FIG. 4 and the like), the test data of the subject, profile information, and possessed device information for each test / medical institution. According to this, the transmission information to be transmitted to the target person is determined (see, for example, the information providing unit 1c in FIG. 1, S11 in FIG. 4 and the like). Therefore, the information can be transmitted to the target person in consideration of the equipment owned by the examination / medical institution. Further, in the present embodiment, it is possible to introduce an effective facility based on the health-related information obtained unconsciously on a daily basis, and it is possible to save the trouble of visiting the facility many times.
　また、本発明の一実施形態においては、対象者の検査データを入力し（例えば、図１の情報判定機器２、図５のＳ２１等参照）、対象者の検査・医療機関への来院・検査・服薬情報に従って学習された推論モデルを有する推論部において、対象者の検査データの変化パターンを入力して推論を行い、この推論結果に基づいて、対象者に伝達する伝達情報を決定している（例えば、図１の情報提供部１ｃ、推論エンジン７、図３（ａ）、図５のＳ３５等参照）。このため、検査データの時系列的変化から、対象者に医療機関等への来院時期を事前に予測することによって知らせることができる。症状が悪化する前に、適切な検査や治療を受けることが可能となる。 Further, in one embodiment of the present invention, the test data of the subject is input (see, for example, the information determination device 2 in FIG. 1, S21 in FIG. 5 and the like), and the test of the subject / visit to a medical institution / inspection. -In the inference unit that has an inference model learned according to the medication information, the change pattern of the test data of the subject is input and inference is performed, and the transmission information to be transmitted to the subject is determined based on the inference result. (For example, refer to the information providing unit 1c of FIG. 1, the inference engine 7, FIG. 3 (a), S35 of FIG. 5 and the like). Therefore, it is possible to notify the subject by predicting the time of visit to a medical institution or the like in advance from the time-series change of the examination data. It will be possible to receive appropriate tests and treatment before the symptoms worsen.
　また、本発明の一実施形態における情報伝達用のプログラムは、対象者の検査データを入力するステップ（例えば、図１の情報判定機器２、図５のＳ２１等参照）、対象者の検査・医療機関への来院・検査・服薬情報に従って学習された推論モデルを有する推論部において、対象者の特定期間にわたる検査データの変化パターンを入力し、この変換パターンを用いて推論を行うステップ、この推論結果に基づいて、対象者に伝達する伝達情報を決定するステップを有している。このため、検査データの時系列的変化から、対象者に医療機関等への来院時期を事前に予測することによって知らせることができる。症状が悪化する前に、適切な検査や治療を受けることが可能となる。 Further, the information transmission program according to the embodiment of the present invention includes a step of inputting test data of a subject (see, for example, information determination device 2 in FIG. 1, S21 in FIG. 5 and the like), and inspection / medical treatment of the subject. In the inference department that has an inference model learned according to the visit / examination / medication information to the institution, the step of inputting the change pattern of the examination data over a specific period of the subject and making an inference using this conversion pattern, this inference result. Has a step of determining the transmitted information to be transmitted to the subject based on. Therefore, it is possible to notify the subject by predicting the time of visit to a medical institution or the like in advance from the time-series change of the examination data. It will be possible to receive appropriate tests and treatment before the symptoms worsen.
　検査結果や検査結果をデータベースに蓄積し、検査結果を利用することが開示されている先行技術はあるが、検査結果に基づいて最適な医療機関等を知らせることについてまでは記載されていない。また、遠隔地の医療関係者に排泄物の観察を依頼することを想定する提案はあるが、最適な医療機関等を知らせることについてまで提案されていない。さらに、ヒストリーブラウザを表示することについての提案もあるが、医療機関等の保有機器に基づいて、被検者の症状を調べるに最適な施設を告知することについてまで提案されていない。多くの一般ユーザは、日々の生活習慣の中で、無意識に携帯端末や家電、バス・トイレ・洗面所などサニタリー設備に設けたセンサ、監視カメラ、見守りカメラ、家庭用の体温計や体重計、体組成計、血圧計等で生体情報を時系列でモニタリングしている場合がある。本発明の一実施形態においては、日々モニタリングし、モニタリングによって得た情報を積極的に活用し、生体情報、検体情報に基づいて、適切な施設で精密検査を受けることができるようなアドバイス等のカスタマイズ情報を提供している。このため、対象者の状況を考慮し、正確な健康状態を把握するための検査ができる施設、および／または治療ができる施設の情報を提供することができる。 Although there is a prior art that discloses that the test results and test results are stored in a database and the test results are used, there is no description about notifying the optimal medical institution based on the test results. In addition, although there is a proposal to ask medical personnel in remote areas to observe excrement, there is no proposal to inform the optimal medical institution. Furthermore, although there is a proposal to display a history browser, there is no proposal to announce the optimal facility for examining the symptoms of a subject based on the equipment owned by a medical institution or the like. Many general users unknowingly install sensors, surveillance cameras, watching cameras, home thermometers, weight scales, and bodies in sanitary facilities such as mobile terminals, home appliances, baths, toilets, and washrooms in their daily lives. Biological information may be monitored in chronological order with a composition meter, sphygmomanometer, or the like. In one embodiment of the present invention, daily monitoring is performed, information obtained by the monitoring is actively utilized, and advice is provided so that a detailed examination can be performed at an appropriate facility based on biological information and sample information. Provides customization information. Therefore, in consideration of the situation of the subject, it is possible to provide information on the facilities where the examination can be performed and / or the facilities where the treatment can be performed in order to grasp the accurate health condition.
　本実施形態の説明に当たっては、情報判定機器２として、トイレに併設された各種センサで検便、採便等を行った結果を有効利用することについて多々説明したが、当然、これに限るものではない。情報判定機器２としては、対象者の健康関連情報、例えば、バイタル情報、検体情報等を取得するための機器であればよい。最も簡単な例ではスマートフォンなどの携帯端末で得られた顔画像情報、それに基づく心拍情報などにも応用でき、これらの情報を活用してもよい。また、ウェアラブル端末などユーザに密着した状態で使用する機器と連携してもよく、例えば、不整脈のような注意すべきデータも、これらの機器で簡単に取得できる。歩行時の加速度センサのパターンなどによっても、足に影響する健康問題の検出が可能である。機器や体調、飲食や生活シーンの状況によって誤差を含みうる単発のデータの解析でなく、複数のデータを含んでいる履歴パターンを用いて解析することによって、疾病の有無や可能性や回復、通院すべき時期などの情報、アドバイス情報等が、高い精度で提供可能となる。これらの情報が低い精度の場合には、ユーザが診察を受けるのが遅れ、必要以上の心配をすることになる。 In the explanation of the present embodiment, many explanations have been made about effectively utilizing the results of stool test, stool collection, etc. by various sensors attached to the toilet as the information determination device 2, but the present invention is not limited to this. .. The information determination device 2 may be any device for acquiring health-related information of the target person, for example, vital information, sample information, and the like. In the simplest example, it can be applied to face image information obtained from a mobile terminal such as a smartphone, heartbeat information based on the face image information, and the like, and these information may be utilized. In addition, it may be linked with a device such as a wearable terminal that is used in close contact with the user, and data to be noted such as arrhythmia can be easily acquired by these devices. It is also possible to detect health problems that affect the feet by the pattern of the acceleration sensor during walking. By analyzing using a history pattern that includes multiple data, instead of analyzing single-shot data that may contain errors depending on the equipment, physical condition, eating and drinking, and living scenes, the presence or absence of illness, possibility, recovery, and outpatient visits Information such as when to do it, advice information, etc. can be provided with high accuracy. If this information is of low accuracy, the user will be delayed in seeing the doctor and will be more worried than necessary.
　多くのこれまでの提案には、このような精度に対する対策が不十分であったが、本実施形態においては精度を考慮し、かつ、対象者の状況を考慮して、対象者が無理なく医療機関等に足を運べる情報を提供できる。正確な健康状態を把握するための検査や治療を受けることができる施設の情報を提供することができるので、ユーザは、この健康把握で、治療を受けたり、生活習慣の改善を行ったりして、より健康的な生活を送ることができる。 Many of the previous proposals did not have sufficient measures against such accuracy, but in this embodiment, the accuracy is taken into consideration, and the subject's situation is taken into consideration, so that the subject can be treated reasonably. It is possible to provide information that can be visited by institutions. Since it is possible to provide information on facilities where tests and treatments can be received to grasp the accurate health condition, users can receive treatment and improve their lifestyles with this health grasp. , You can lead a healthier life.
　なお、本発明の一実施形態においては、制御部１は、ＣＰＵ、メモリ、ＨＤＤ等から構成されているＩＴ機器として説明した。しかし、ＣＰＵとプログラムによってソフトウエア的に構成する以外にも、各部の一部または全部をハードウエア回路で構成してもよく、ヴェリログ（Verilog）によって記述されたプログラム言語に基づいて生成されたゲート回路等のハードウエア構成でもよく、またＤＳＰ（Digital Signal Processor）等のソフトを利用したハードウエア構成を利用してもよい。これらは適宜組み合わせてもよいことは勿論である。また、ＣＰＵに限らず、コントローラとしての機能を果たす素子であればよく、上述した各部の処理は、ハードウエアとして構成された１つ以上のプロセッサが行ってもよい。例えば、各部は、それぞれが電子回路として構成されたプロセッサであっても構わないし、ＦＰＧＡ（Field Programmable Gate Array）等の集積回路で構成されたプロセッサにおける各回路部であってもよい。または、１つ以上のＣＰＵで構成されるプロセッサが、記録媒体に記録されたコンピュータプログラムを読み込んで実行することによって、各部としての機能を実行しても構わない。 In one embodiment of the present invention, the control unit 1 has been described as an IT device composed of a CPU, a memory, an HDD, and the like. However, in addition to being configured in software by a CPU and a program, part or all of each part may be configured in a hardware circuit, and a gate generated based on the program language described by Verilog. A hardware configuration such as a circuit may be used, or a hardware configuration using software such as a DSP (Digital Signal Processor) may be used. Of course, these may be combined as appropriate. Further, the element is not limited to the CPU, and may be any element that functions as a controller, and the processing of each part described above may be performed by one or more processors configured as hardware. For example, each part may be a processor each of which is configured as an electronic circuit, or may be each circuit part of a processor composed of an integrated circuit such as an FPGA (Field Programmable Gate Array). Alternatively, a processor composed of one or more CPUs may execute the functions of each unit by reading and executing the computer program recorded on the recording medium.
　また、本明細書において説明した技術のうち、主にフローチャートで説明した制御に関しては、プログラムで設定可能であることが多く、記録媒体や記録部に収められる場合もある。この記録媒体、記録部への記録の仕方は、製品出荷時に記録してもよく、配布された記録媒体を利用してもよく、インターネットを通じてダウンロードしたものでもよい。 In addition, among the techniques described in the present specification, the controls mainly described in the flowchart can often be set by a program, and may be stored in a recording medium or a recording unit. The recording method to the recording medium and the recording unit may be recorded at the time of product shipment, the distributed recording medium may be used, or may be downloaded via the Internet.
　また、本発明の一実施形態においては、フローチャートを用いて、本実施形態における動作を説明したが、処理手順は、順番を変えてもよく、また、いずれかのステップを省略してもよく、ステップを追加してもよく、さらに各ステップ内における具体的な処理内容を変更してもよい。 Further, in one embodiment of the present invention, the operation in the present embodiment has been described using a flowchart, but the order of the processing procedures may be changed, or any step may be omitted. Steps may be added, and specific processing contents in each step may be changed.
　また、特許請求の範囲、明細書、および図面中の動作フローに関して、便宜上「まず」、「次に」等の順番を表現する言葉を用いて説明したとしても、特に説明していない箇所では、この順で実施することが必須であることを意味するものではない。 In addition, even if the scope of claims, the specification, and the operation flow in the drawings are explained using words expressing the order such as "first" and "next" for convenience, in the parts not particularly explained, It does not mean that it is essential to carry out in this order.
　本発明は、上記実施形態にそのまま限定されるものではなく、実施段階ではその要旨を逸脱しない範囲で構成要素を変形して具体化できる。また、上記実施形態に開示されている複数の構成要素の適宜な組み合わせによって、種々の発明を形成できる。例えば、実施形態に示される全構成要素の幾つかの構成要素を削除してもよい。さらに、異なる実施形態にわたる構成要素を適宜組み合わせてもよい。 The present invention is not limited to the above embodiment as it is, and at the implementation stage, the components can be modified and embodied within a range that does not deviate from the gist thereof. In addition, various inventions can be formed by an appropriate combination of the plurality of components disclosed in the above-described embodiment. For example, some components of all the components shown in the embodiment may be deleted. In addition, components across different embodiments may be combined as appropriate.
１・・・制御部、１ａ・・・通信制御部、１ｂ・・・ＩＤ判定部、１ｃ・・・情報提供部、１ｄ・・・推論モデル仕様決定部、１ｅ・・・推論依頼部、１ｆ・・・検索部、２・・・情報判定機器、４・・・端末、５・・・学習部、５ａ・・・入出力モデル化部、７・・・推論エンジン、８・・・ＤＢ部、８ａ・・・施設別保有機器一覧、８ｂ・・・施設別ＩＤと来院情報 1 ... Control unit, 1a ... Communication control unit, 1b ... ID determination unit, 1c ... Information provision unit, 1d ... Inference model specification determination unit, 1e ... Inference request unit, 1f ... Search unit, 2 ... Information judgment device, 4 ... Terminal, 5 ... Learning unit, 5a ... Input / output modeling unit, 7 ... Inference engine, 8 ... DB unit , 8a ・ ・ ・ List of owned equipment by facility, 8b ・ ・ ・ ID and visit information by facility

Claims (8)

  1. 　特定の仕様の検査機器を用いて特定期間の時系列的に検査データを取得し、該検査データの変化パターン情報を取得し、上記変化パターン情報に対し、上記特定期間内で上記変化パターンが変化した時点に対応する時点における影響イベント情報を検出し、上記検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得し、
    　特定の人物の上記特定の仕様の検査機器を用いて上記特定期間の幅と類似の期間で時系列的に得た検査データを、上記推論モデルに入力して影響イベントを推論し、該推論の結果を伝達する、
    　ことを特徴とする情報伝達方法。     Inspection data is acquired in a time series of a specific period using an inspection device having a specific specification, change pattern information of the inspection data is acquired, and the change pattern changes with respect to the change pattern information within the specific period. An inference model was obtained by detecting the impact event information at the time corresponding to the time point and learning the result of annotating the detected impact event information as teacher data.
    Inspection data obtained in time series in a period similar to the width of the specific period using the inspection device of the specific specification of a specific person is input to the inference model to infer the influence event, and the inference is performed. Communicate the results,
    An information transmission method characterized by the fact that.
  2. 　特定の健康対応イベントの入力に対応し、上記健康対応イベントに先立って特定の仕様の検査機器を用いて時系列的に検査データを取得し記録されていた、該検査データの変化パターン情報に対し、上記健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報をアノテーションした結果を教師データとして学習させた推論モデルを取得し、
    　特定の人物の上記特定の仕様の検査機器を用いて時系列的に得た検査データを、上記推論モデルに入力して設備、および／または備品、および／または環境の情報を推論し、該推論結果から上記特定の人物用にカスタマイズした健康対応イベントを、上記特定の人物に伝達する、
    　ことを特徴とする情報伝達方法。     For the change pattern information of the test data, which corresponds to the input of the specific health response event and the test data is acquired and recorded in time series using the test device of the specific specifications prior to the above health response event. , Acquire an inference model in which the result of annotating the equipment and / or equipment and / or environment information at the time of the above health response event is trained as teacher data.
    Inspection data obtained in time series using the inspection equipment of the specific specifications of a specific person is input to the inference model to infer information on equipment and / or equipment and / or environment, and the inference is made. From the results, the health response event customized for the specific person is transmitted to the specific person.
    An information transmission method characterized by the fact that.
  3. 　特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得して作成した上記検査データの変化パターン情報と、上記特定の期間内における上記検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録可能なデータベースを作成し、
    　上記データベースを用いて、上記変化パターンのトレンド変化と上記イベントのタイミングの関係に基づいて、上記イベント情報のうち、上記トレンド変化に影響したイベントの情報提供を可能とする、
    　ことを特徴とする情報伝達方法。     The change pattern information of the inspection data created by acquiring a plurality of inspection data in a time series using an inspection device having a specific specification, and the acquisition source of the inspection data within the specific period. Create a database that can record the timing of each of multiple event information related to
    Using the database, it is possible to provide information on events that affect the trend change among the event information based on the relationship between the trend change of the change pattern and the timing of the event.
    An information transmission method characterized by the fact that.
  4. 　上記イベント情報を構成毎に分解して構成要素を抽出し、上記トレンド変化に影響したイベント情報を抽出した上記構成要素に基づいてカスタマイズして情報提供することを特徴とする請求項３に記載の情報伝達方法。　 The third aspect of claim 3, wherein the event information is decomposed for each component, the components are extracted, and the event information affecting the trend change is customized based on the extracted components to provide the information. Information transmission method.
  5. 　上記データベースを用いて、上記トレンド変化に影響したイベントを提供するための　推論モデルは、検査データを取得し、該検査データの変化パターンを学習用推論部の入力とし、出力すべきアドバイスをアノテーション情報として、学習することによって推論モデルを生成し、
    　この生成した推論モデルを用い、上記対象者の変化パターン情報を上記推論モデルに入力することにより推論結果を得て、この得られた推論結果に基づいて、上記伝達情報を決定することを特徴とする請求項４に記載の情報伝達方法。     The inference model for providing the event affecting the trend change using the above database acquires the inspection data, uses the change pattern of the inspection data as the input of the learning inference unit, and annotates the advice to be output. Generate an inference model by learning as
    Using this generated inference model, the inference result is obtained by inputting the change pattern information of the target person into the inference model, and the transmission information is determined based on the obtained inference result. The information transmission method according to claim 4.
  6. 　特定の仕様の検査機器を用いて特定期間の時系列的に検査データを取得し、該検査データの変化パターン情報を取得するデータ取得部と、
    　上記変化パターン情報に対し、上記特定期間内で上記変化パターンが変化した時点に対応する時点における影響イベント情報を検出し、上記検出された影響イベント情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する学習部と、
    　特定の人物の上記特定の仕様の検査機器を用いて上記特定期間の幅と類似の期間で時系列的に得た検査データを、上記推論モデルに入力して影響イベントを推論する推論部と、
    　上記推論の結果を伝達する情報伝達部と、
    　を有することを特徴とする情報伝達装置。     A data acquisition unit that acquires inspection data in chronological order for a specific period using an inspection device with specific specifications and acquires change pattern information of the inspection data.
    Inference that the effect event information at the time corresponding to the time when the change pattern changes within the specific period is detected for the change pattern information, and the result of annotating the detected effect event information is learned as teacher data. The learning department that acquires the model and
    An inference unit that infers an influence event by inputting inspection data obtained in time series in a period similar to the width of the specific period using the inspection device of the specific specification of a specific person into the inference model.
    An information transmission unit that conveys the results of the above inference,
    An information transmission device characterized by having.
  7. 　特定の健康対応イベントの入力に対応し、上記健康対応イベントに先立って特定の仕様の検査機器を用いて時系列的に検査データを取得し記録するデータ取得部と、
    　記録されていた上記検査データについての変化パターン情報に対し、上記健康対応イベントの行われた時の設備、および／または備品、および／または環境の情報をアノテーションした結果を教師データとして学習させた推論モデルを取得する学習部と、
    　特定の人物の上記特定の仕様の検査機器を用いて時系列的に得た検査データを、上記推論モデルに入力して設備、および／または備品、および／または環境の情報を推論する推論部と、
    　上記推論部による推論結果から上記特定の人物用にカスタマイズした健康対応イベントを、上記特定の人物に伝達する情報伝達部と、
    　を有することを特徴とする情報伝達装置。     A data acquisition unit that responds to the input of a specific health response event and acquires and records inspection data in chronological order using an inspection device with specific specifications prior to the above health response event.
    Inference that the result of annotating the equipment, / or equipment, and / or environment information at the time of the health response event was learned as teacher data for the recorded change pattern information of the test data. The learning department to acquire the model and
    An inference unit that inputs inspection data obtained in time series using the inspection equipment of the above-mentioned specific specifications of a specific person into the above-mentioned inference model to infer information on equipment and / or equipment and / or environment. ,
    The information transmission unit that transmits the health response event customized for the specific person from the inference result by the inference unit to the specific person, and the information transmission unit.
    An information transmission device characterized by having.
  8. 　特定の仕様の検査機器を用いて特定の期間に亘って時系列的に複数の検査データを取得するデータ取得部と、
    　上記検査データを取得して作成した上記検査データの変化パターン情報と、上記特定の期間内における上記検査データの取得元に関する複数のイベント情報のそれぞれのタイミングと、を関連付けて記録可能なデータベースを作成するデータ作成部と、
    　上記データベースを用いて、上記変化パターンのトレンド変化と上記イベントのタイミングの関係に基づいて、上記イベント情報のうち、上記トレンド変化に影響したイベントを抽出して情報提供を可能とする情報提供部と、
    　を有することを特徴とする情報伝達装置。     A data acquisition unit that acquires multiple inspection data in chronological order over a specific period using inspection equipment with specific specifications,
    Create a database that can record the change pattern information of the inspection data created by acquiring the inspection data and the timing of each of the plurality of event information related to the acquisition source of the inspection data within the specific period. Data creation department and
    Using the database, based on the relationship between the trend change of the change pattern and the timing of the event, the information providing unit that can extract the event that affected the trend change from the event information and provide the information. ,
    An information transmission device characterized by having.
PCT/JP2020/041125 2020-01-10 2020-11-02 Information transmitting device and information transmitting method WO2021140731A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202080085581.7A CN114868203A (en) 2020-01-10 2020-11-02 Information transmission device and information transmission method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
PCT/JP2020/000763 WO2021140670A1 (en) 2020-01-10 2020-01-10 Information transmission device and information transmission method
JPPCT/JP2020/000763 2020-01-10

Publications (1)

Publication Number Publication Date
WO2021140731A1 true WO2021140731A1 (en) 2021-07-15

Family

ID=76787834

Family Applications (2)

Application Number Title Priority Date Filing Date
PCT/JP2020/000763 WO2021140670A1 (en) 2020-01-10 2020-01-10 Information transmission device and information transmission method
PCT/JP2020/041125 WO2021140731A1 (en) 2020-01-10 2020-11-02 Information transmitting device and information transmitting method

Family Applications Before (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/000763 WO2021140670A1 (en) 2020-01-10 2020-01-10 Information transmission device and information transmission method

Country Status (2)

Country Link
CN (1) CN114868203A (en)
WO (2) WO2021140670A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023157596A1 (en) * 2022-02-15 2023-08-24 ソニーグループ株式会社 Information processing method, information processing device, program, and information processing system

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116206774B (en) * 2023-04-27 2023-07-14 深圳市浩然盈科通讯科技有限公司 Method and system for automatically matching nursing treatment scheme by combining big data

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009080700A (en) * 2007-09-26 2009-04-16 Toshiba Corp Health/medical information management analysis device
JP2016218536A (en) * 2015-05-15 2016-12-22 コニカミノルタ株式会社 Medical examination support system, server device, and program
WO2017037946A1 (en) * 2015-09-04 2017-03-09 株式会社日立システムズ Lifestyle management assistance service system and method
JP2018147407A (en) * 2017-03-08 2018-09-20 オムロン株式会社 Factor estimation device, factor estimation system, and factor estimation method
JP6566409B1 (en) * 2018-08-24 2019-08-28 株式会社鈴康 Information processing apparatus, program, and information processing method

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1189822A (en) * 1997-09-16 1999-04-06 Forward Inku:Kk Method for blood examination at home
US20040122704A1 (en) * 2002-12-18 2004-06-24 Sabol John M. Integrated medical knowledge base interface system and method
JP5631635B2 (en) * 2010-06-03 2014-11-26 オリンパスメディカルシステムズ株式会社 Medical work support device
JP6494935B2 (en) * 2014-07-09 2019-04-03 キヤノンメディカルシステムズ株式会社 Inspection information provider
MX2017012732A (en) * 2015-04-08 2017-11-30 Koninklijke Philips Nv System for laboratory values automated analysis and risk notification in intensive care unit.
EP3660752A4 (en) * 2017-07-24 2021-04-14 Axion Research Inc. Assistance system for estimating internal state of system-of-interest
JP6804489B2 (en) * 2018-05-31 2020-12-23 キユーピー株式会社 Information providing device and information providing method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009080700A (en) * 2007-09-26 2009-04-16 Toshiba Corp Health/medical information management analysis device
JP2016218536A (en) * 2015-05-15 2016-12-22 コニカミノルタ株式会社 Medical examination support system, server device, and program
WO2017037946A1 (en) * 2015-09-04 2017-03-09 株式会社日立システムズ Lifestyle management assistance service system and method
JP2018147407A (en) * 2017-03-08 2018-09-20 オムロン株式会社 Factor estimation device, factor estimation system, and factor estimation method
JP6566409B1 (en) * 2018-08-24 2019-08-28 株式会社鈴康 Information processing apparatus, program, and information processing method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023157596A1 (en) * 2022-02-15 2023-08-24 ソニーグループ株式会社 Information processing method, information processing device, program, and information processing system

Also Published As

Publication number Publication date
WO2021140670A1 (en) 2021-07-15
CN114868203A (en) 2022-08-05

Similar Documents

Publication Publication Date Title
CN105260588B (en) A kind of health guards robot system and its data processing method
National Academies of Sciences, Engineering, and Medicine Improving diagnosis in health care
JP6700283B2 (en) A medical differential diagnosis device adapted to determine an optimum sequence of diagnostic tests for identifying a disease state by adopting a diagnostic validity standard
Peleg et al. MobiGuide: a personalized and patient-centric decision-support system and its evaluation in the atrial fibrillation and gestational diabetes domains
Rothman et al. Clinical implications and validity of nursing assessments: a longitudinal measure of patient condition from analysis of the Electronic Medical Record
JP7190859B2 (en) Health information processing method, health information processing device, computer program, and learning model
KR101141425B1 (en) Method of personalized health care and treatment using on-line information processing system and server device for online health care and medical service
WO2005122033A1 (en) Medical total information apparatus and medical total information system
US20060112050A1 (en) Systems and methods for adaptive medical decision support
US20190287661A1 (en) Related systems and method for correlating medical data and diagnostic and health treatment follow-up conditions of patients monitored in real-time
KR100737382B1 (en) Operaton method for medical diagnosis data
EP3547320A2 (en) Related systems and method for correlating medical data and diagnostic and health treatment follow-up conditions of patients monitored in real-time
WO2021140731A1 (en) Information transmitting device and information transmitting method
JP2013191021A (en) Physical examination information providing device and physical examination information providing method
JP2020135401A (en) Software and diagnosis support device
AU7719501A (en) Online medical evaluation and treatment system, method and portal
US11322250B1 (en) Intelligent medical care path systems and methods
KR20070106942A (en) Method for service of anticipatory medical diagnosis
WO2021152710A1 (en) Information transmission device and information transmission method
US20090070145A1 (en) Method and system for coronary artery disease care
WO2021176622A1 (en) Sensor determination device and sensor determination method
WO2021181634A1 (en) Teacher data collection requesting device, and teacher data collection method
Hine et al. Decision making by emergency room physicians and residents: implications for the design of clinical decision support systems
WO2021171596A1 (en) Inference system, learning request device, and learning request method
Sabbir et al. Web-based health monitoring system and textual mining

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20912926

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20912926

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP