WO2019167406A1 - Disease-warning-information providing support system - Google Patents

Disease-warning-information providing support system Download PDF

Info

Publication number
WO2019167406A1
WO2019167406A1 PCT/JP2018/047678 JP2018047678W WO2019167406A1 WO 2019167406 A1 WO2019167406 A1 WO 2019167406A1 JP 2018047678 W JP2018047678 W JP 2018047678W WO 2019167406 A1 WO2019167406 A1 WO 2019167406A1
Authority
WO
WIPO (PCT)
Prior art keywords
disease
information
onset
probability
person
Prior art date
Application number
PCT/JP2018/047678
Other languages
French (fr)
Japanese (ja)
Inventor
英之 明石
Original Assignee
メドケア株式会社
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 メドケア株式会社 filed Critical メドケア株式会社
Publication of WO2019167406A1 publication Critical patent/WO2019167406A1/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
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • 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 a disease attention information provision support system.
  • the contents described in the insured's receipt and the result of the medical examination are used to determine the insured person.
  • Extract the name of the diagnosis and the test result data of the health check related to the name of the diagnosis compare the extracted test result data with the medical guidelines for the name of the diagnosis, and provide an index indicating the degree of disease progression of the diagnosis
  • the calculation is performed, and it is determined whether or not the insured needs to have a medical examination at a medical institution based on the calculation result.
  • the result of this determination can be used when a doctor or the like gives instructions to the insured such as recommending continued medical care at a medical institution.
  • this type of information can be notified to each insured person as it is.
  • Patent Document 1 the information obtained by the system disclosed in Patent Document 1 is a person who has specialized knowledge such as the name of injury and illness, numerical values of test results such as blood glucose level and total cholesterol level, and “continuation of consultation” based on the numerical values. It is information on the assumption that has been viewed.
  • the purpose of the present invention is to allow a person who has developed or has a possibility of developing a disease to fully recognize the necessity of improvement in view of the problems of the prior art, and through medical examinations and medical examinations. It is to provide a disease attention information providing support system that can promote improvement of health condition.
  • the disease attention information provision support system of the present invention is A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons; An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information; Attention information generation unit for generating disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or a plurality of diseases digitized by the onset probability digitization unit; Information on the health status of the subject at the first time is input to the model including the information including the health status information of any man at any time as an input and information indicating the health status of the man at a time after the time is output.
  • the attention information generation unit is configured to generate the disease attention information based on a numerical value of the onset possibility at the second time point for each disease.
  • the onset probability quantification unit first quantifies the onset probability for each person's disease based on the health condition information including information on the health condition of one or more persons.
  • the attention information generation unit generates disease attention information including information indicating the numerical value of the onset probability of at least one of the one or more diseases digitized by the onset probability digitization unit.
  • disease attention information including information indicating the numerical value of the probability of developing at least one disease among one or more diseases is generated. Is done.
  • the disease attention information provision support system of the present invention predicts the target person at a second time point after the first time point using information including the health condition information of the target person at the first time point.
  • a prediction unit is provided for acquiring information indicating a health condition.
  • the prediction unit uses information including information on the health status of any person at any time as an input, and outputs information indicating the health status of the person at a time later than that time. Information indicating the predicted health state of the subject person at the second time point is acquired.
  • the onset probability digitization unit quantifies the onset probability at the second time point for each person's disease based on the information acquired by the prediction unit.
  • the attention information generation unit generates disease attention information based on a numerical value of the possibility of onset at the second time point for each disease.
  • the disease attention information includes improvement advice information that is information on a method for improving the possibility of onset for each disease, It is preferable that the attention information generation unit is configured to generate the improvement advice information only for a disease having a numerical value of a disease onset probability equal to or greater than a certain value.
  • the disease attention information includes improvement advice information that is information related to a method for improving the possibility of onset for each disease.
  • the improvement advice information is generated only for a disease whose numerical value of the probability of developing the disease is a certain value or more.
  • the improvement advice information is generated only for the disease for which the numerical value of the onset of the disease is a certain value or more, so that the disease having a high priority for improvement can be more strongly impressed.
  • the disease attention information is configured to include information on complications that may occur due to a disease that each person may develop according to a numerical value of the possibility of developing the disease. preferable.
  • the disease attention information includes information on complications that may develop due to the diseases that each person may develop, so that the disease has or has developed.
  • the potential for improvement can be impressed by those who have the potential more than when information is provided about individual diseases.
  • the disease attention information includes a graph representing a numerical value of the likelihood of onset for each person's disease, and the graph is arranged in descending order of the numerical value of the probability of onset for each disease.
  • the disease attention information includes a graph representing the numerical value of the onset possibility for each person's disease. Moreover, the graph is arranged in descending order of the numerical value of the onset probability for each disease.
  • the disease attention information is configured to include information related to the ranking of the probability of occurrence of each person's disease in the group formed by all or part of the plurality of persons.
  • the disease attention information includes information related to the ranking of the onset probability of each person's disease in a group composed of all or part of the plurality of persons.
  • the onset probability quantification unit extracts health state information correlated with the onset probability for each disease from the health state information, and determines the onset probability for each person's disease based on the correlated health state information. It is preferable to be configured to be numerical.
  • the possibility of onset for each disease is quantified based on the information. Numerical processing can be performed more efficiently than considering all items of state information.
  • the prediction unit includes information including health state information at each time point of the plurality of people, health state information at each time point of the plurality of people, and time points before the respective time points of the plurality of people. It is preferable that the model is generated or updated using information including the magnitude of change between the health status information.
  • the prediction unit includes information including health state information at a certain time point of each of the plurality of persons, health state information of each of the plurality of persons, and each time point of the plurality of persons.
  • information including health state information at a certain time point of each of the plurality of persons, health state information of each of the plurality of persons, and each time point of the plurality of persons.
  • the prediction unit updates the model using information including the magnitude of the change in the actual health status information of a plurality of people, the information indicating the predicted health status of the target person at the second time point Accuracy can be increased.
  • the onset probability quantification unit repeatedly quantifies the onset probability of each person's disease, and the attention information generation unit generates disease attention information for a person whose value has changed at least a certain level. It is preferable that it is comprised.
  • the onset probability quantification unit repeatedly quantifies the onset probability for each person's disease, and the attention information generation unit generates the disease attention information, so the disease attention information is generated only once. As compared with the case, it is possible to make a person who has developed or has a possibility of developing a disease more strongly aware of the need for improvement.
  • the disease attention information is generated for people who have at least a certain change in the likelihood of developing the disease, for example, the risk of developing the disease is increased for those who have a sudden increase in the likelihood of developing the disease. It is possible to reliably recognize the situation where the nature is rapidly increasing.
  • the attention information generation unit is configured to vary the frequency of generating the disease attention information according to the numerical value of the possibility of onset for each person's disease.
  • the attention information generation unit changes the frequency of generating the disease attention information according to the numerical value of the possibility of onset for each person's disease.
  • the disease attention information is generated frequently for people who want to have an early visit at a medical institution, and for those who do not have a high possibility of onset at a certain interval.
  • the disease attention information provision support system of the present invention is A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons; An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information; An attention information generation unit that generates disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or more diseases digitized by the onset probability digitization unit,
  • the disease attention information is configured to include information indicating a rank in a group composed of all or a part of the plurality of persons with respect to the possibility of onset for each disease of a person to whom the disease attention information is provided. It is characterized by being.
  • the disease attention information is information indicating a rank in a group composed of all or a part of the plurality of persons, with respect to the possibility of occurrence of each disease of the target person to whom the disease attention information is provided. Is included.
  • each person In order to respond to such issues, it is possible to provide each person with statistical information that can list the numerical value of the likelihood of developing each disease, including other people. If the specific order of the onset probability for each person's disease cannot be intuitively understood, there is a concern that each person who does not have expertise cannot fully understand the seriousness of the medical condition.
  • the disease attention information is composed of “all or part of a plurality of people” regarding the possibility of occurrence of each disease of “the person to whom the disease attention information is provided”. Since the target person who receives the information can intuitively understand the specific order of the onset probability for each disease.
  • the target person who receives the provision of the disease attention information can intuitively understand the specific ranking of the onset probability for each disease. Have or will develop a disease rather than providing statistical information that allows you to list only the probability of onset or the number of possible onset for each disease, including other people Can make it easy to recognize the high necessity for improvement.
  • the disease attention information provision support system of the present invention is A health condition storage unit that stores health condition information including values of a plurality of items relating to the health condition of one or more people; An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information; An attention information generation unit that generates disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or more diseases digitized by the onset probability digitization unit, The onset probability quantification unit identifies the item of the health status information correlated with the onset probability for each disease by analyzing the health status information of one or a plurality of people, and the health status of the target person It is configured to extract a value of an item correlated with the onset possibility for each specified disease from the information, and to quantify the onset probability for each person's disease based on the value of the extracted item. It is characterized by.
  • the onset probability quantification unit identifies items of the health condition information correlated with the onset possibility for each disease by analyzing the health condition information of one or a plurality of persons.
  • the onset probability quantification unit extracts the value of the item correlated with the onset probability for each of the identified diseases from the health status information of the target person.
  • the onset probability quantification unit is configured to quantify the onset probability for each person's disease based on the value of the extracted item.
  • the person in charge may specify items that have a correlation between these items and the likelihood of onset for each disease by visual inspection or manual calculation. In the first place, it is expected to be very difficult.
  • the onset probability quantification unit performs the “analysis of the health status information of one or a plurality of persons” to thereby “correlate with the onset probability for each disease. “Identify items of health condition information” and “Extract values of items correlated with the probability of occurrence of each identified disease from the health condition information of the target person” Based on the above, it is configured to ⁇ numerize the probability of occurrence of each person's disease '', so consider the value of items that are actually correlated with the probability of occurrence of each disease, In addition to being numerical, it is possible to efficiently quantify the likelihood of onset rather than considering the values of all items of health condition information.
  • the disease attention information provision support system of the present invention is A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons; An onset probability quantification unit that quantifies the onset probability of each of a plurality of diseases of each person based on the stored health condition information; An attention information generation unit that generates disease attention information including information indicating a numerical value of the onset probability of at least one disease among a plurality of diseases digitized by the onset probability digitization unit, The attention information generating unit recognizes a disease whose numerical value of the probability of occurrence satisfies a predetermined condition among the plurality of diseases, recognizes a complication stored in association with the disease, and stores information on the complication It is comprised so that the said disease caution information containing may be produced
  • the disease attention information provision support system of the present configuration “recognizes a disease in which the numerical value of the probability of occurrence among the plurality of diseases satisfies a predetermined condition” and further stores “a complication associated with the disease and stored. And “generates the disease attention information including information related to the complication”.
  • the disease attention information includes information on complications that may develop due to the diseases that each person may develop.
  • the need for improvement can be impressed more strongly by those who have the information than when information about individual diseases is provided.
  • the disease attention information provision support system of this configuration it is possible to impress the necessity of improving the possibility of developing a disease more strongly than when information is provided about individual diseases, and for each person. It is possible to provide information on complications according to the numerical value of the probability of disease occurrence.
  • the flowchart of the process from a health condition prediction process to the output of disease caution information The flowchart of a preparation process.
  • the flowchart of onset possibility correlation item extraction processing The flowchart of onset possibility calculation model derivation processing.
  • the flowchart of a change correlation item extraction process The flowchart of a 2nd time information acquisition model derivation process.
  • the flowchart of a health condition prediction process The figure which shows an example of the processing content which acquires the information which shows the healthy state of the object person's prediction in the 2nd time point by a prediction part.
  • the flowchart of the onset possibility numerical conversion process The figure which shows an example of the processing content which digitizes the onset possibility for every disease of each person by the onset possibility digitization part.
  • the flowchart of attention information generation processing The figure which shows the example of the conversion table of the numerical value of the onset possibility of a disease, and the classification of onset possibility.
  • the disease attention information includes information on the possibility of developing the disease according to the numerical value of the possibility of developing the disease of each person 40 and the numerical value of the possibility of developing the disease, which are generated in the attention information generation process (STEP 80) described later. Information.
  • the disease attention information is printed on a paper medium 32 by the terminal 30 and provided to each person 40 as shown in FIG.
  • the disease warning information may be provided by being output by the terminal 30 as a document file including the contents shown in FIG. 1 and downloaded to the terminal 41 such as a personal computer, tablet, or smartphone used by each person 40. Good.
  • the disease attention information includes, for example, the past and most recently acquired health state information 321 of each person 40, improvement advice information 322 that is information on how to improve the possibility of developing each disease, disease Information 323 relating to complications that may occur due to a disease that each person may develop according to the numerical value of the possibility of onset, a graph 324 that represents the numerical value of the probability of occurrence for each disease, It includes information 325 relating to the ranking of the probability of occurrence of each person's disease in a group composed of all or part of the person, and a graph 326 including numerical values of the probability of occurrence at the second time point for each disease.
  • the disease caution information does not necessarily include all the information, and may include a part.
  • a specific method for generating disease attention information will be described later.
  • the disease attention information provision support system is a system for generating disease attention information including information on the possibility of developing a disease for one or a plurality of people.
  • the disease attention information provision support system includes a disease attention information provision support server 10 and one or more terminals 30.
  • the disease attention information provision support server 10 and the one or more terminals 30 are configured to be able to communicate with each other via an information communication network 20 such as a LAN or the Internet.
  • an information communication network 20 such as a LAN or the Internet.
  • FIG. 2 one terminal 30 is shown.
  • the disease attention information provision support server 10 may operate as the terminal 30.
  • the disease attention information provision support system does not include the terminal 30.
  • the disease attention information provision support server 10 includes a server control unit 100 and a server storage unit 110. A part or all of the computer constituting the disease caution information provision support server 10 may be configured by a computer constituting the terminal 30.
  • the server control unit 100 includes an arithmetic processing device such as a CPU (Central Processing Unit), a main storage device, and an input / output device.
  • the server control unit 100 is configured by one or a plurality of processors.
  • the server control unit 100 functions as an onset probability digitizing unit 101, a predicting unit 102, a caution information generating unit 103, and a caution information transmitting unit 104 that execute a calculation process described later by reading and executing a predetermined program. .
  • the server storage unit 110 includes, for example, a storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), or the like.
  • a storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), or the like.
  • the server storage unit 110 is configured to store the calculation result of the server control unit 100 or the health condition information captured from the terminal 30.
  • the server storage unit 110 includes a health state storage unit 111, an onset probability correlation item storage unit 112, a calculation model storage unit 113, a change correlation item storage unit 114, a prediction model storage unit 115, a predicted value information storage unit 116, and an onset possibility.
  • a numerical information storage unit 117, a history storage unit 118, and a conversion table storage unit 119 are provided.
  • the health status storage unit 111 stores health status information that is information including information on the health status of one or more people.
  • the health status storage unit 111 includes an ID for identifying each person 40, the date of acquisition of health status information, age, sex, height, weight, waist circumference, total cholesterol, ⁇ -GT, One or more sets of information on items related to the health status of each person 40 and information associated therewith composed of uric acid levels, blood glucose levels, HbA1c, blood pressure, smoking habits, exercise habits, drinking habits, and disease names being treated Storing.
  • the health condition storage unit 111 can store a plurality of sets of the information for different acquisition dates for the same ID.
  • the health state storage unit 111 can store an information set having a certain interval (for example, one year) with a date of acquisition.
  • This one year is not limited to one year in a strict sense, but may be a period with a certain range, for example, 10 months to 14 months, or an information set having different acquisition years or acquisition years. It may be.
  • the information set acquired on the most recent acquisition date among the set of the information for the same ID is referred to as “latest health status information”, and the information acquired on the second new acquisition date.
  • An information set that is included in a set period before the acquisition date of the latest health condition information (for example, 10 months to 14 months before) is set as “health condition information for one year ago”
  • the information set acquired on the third most recent acquisition date, and the acquisition date is one year before the specified period (for example, 10 to 14 months before).
  • the information set included in the range is referred to as “health state information two years ago”.
  • the present invention is not limited thereto, and any information set may be used as long as the acquisition date has a certain interval. .
  • the number of information sets with different acquisition times is not limited to three, and may be two or four or more.
  • the acquisition time of the information set may be the same or different.
  • information sets with different acquisition times for a single person are the latest, 1 year ago, 2 years ago health status information, 1 year ago, 2 years ago and 3 years ago health status information, A plurality may be used.
  • the health status storage unit 111 receives health status information from a terminal 30 such as a hospital or a health insurance association via an external storage medium 31 such as a CD-ROM, DVD-ROM, or USB memory, or via the information communication network 20. Capture.
  • the onset probability correlation item storage unit 112 stores one or a plurality of sets of items related to the health status correlated with the disease name and the onset possibility of the disease.
  • the calculation model storage unit 113 stores a model for calculating the onset probability for each disease derived by the onset probability digitizing unit 101 in the onset possibility calculation model preparation process (STEP 402) described later.
  • the change correlation item storage unit 114 stores a set of items of health state information correlated with a part or all of the item names of the health state information and the magnitude of change in the value of each item. One or more are stored.
  • the prediction model storage unit 115 includes information including the health state information at the first time point, the health state information at the first time point, and the information derived by the prediction unit 102 in the second time point information acquisition model preparation process (STEP 404) described later.
  • a model that outputs information that can specify a health state at a second time point after the first time point by inputting information including the magnitude of change between the health state information at a time point before the first time point and 1 Or a plurality are stored.
  • the first time point is the date of acquisition of the latest health information of each person 40
  • the second time point is, for example, one year after the first time point and before the first time point. Is the date of acquisition of the health status information of each person 40 one year ago.
  • the predicted value information storage unit 116 is a numerical value indicating the health status of each person 40 at the second time point calculated by the prediction unit 102 in the health status prediction process (STEP 50) described later. And one or more sets of information associated therewith.
  • the predicted value information storage unit 116 stores a numerical value indicating the health status of the third time point that is one year after the second time point for the same ID, and the date of the fourth time point that is one year later. It is possible to store a plurality of numerical values indicating health state information in the future date after the second time point, such as a numerical value indicating the health state in.
  • the onset possibility numerical value information storage unit 117 is calculated or set by the onset possibility number conversion unit 101 in the onset possibility numerical value calculation process (STEP 601), such as diabetes risk, dyslipidemia risk, Information that classifies the type according to the numerical value of the onset probability for each person's 40 disease, such as the numerical value of the onset possibility for each person's 40 disease, the diabetes risk type, the dyslipidemia risk type, and the like.
  • STEP 601 onset possibility numerical value calculation process
  • the onset probability numerical information storage unit 117 stores one or more sets of the information.
  • the onset probability numerical information storage unit 117 can store a plurality of sets of the information for different acquisition dates for the same ID.
  • the history storage unit 118 sets one or more sets of the caution information last generation date and the information associated therewith as the date of generation of the latest disease caution information for each person 40. Multiple items are stored.
  • the conversion table storage unit 119 stores conversion tables to be referred to when the disease attention information provision support server 10 executes various processes.
  • the terminal 30 is configured by a desktop computer, a tablet terminal, a smartphone, or the like.
  • the terminal 30 includes a storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), etc., and a server for providing health status information and disease attention information for one or more people.
  • the disease attention information received from 10 is stored.
  • the terminal 30 prints on a paper medium 32 and provides disease warning information to each person 40.
  • the terminal 30 outputs a document file including the contents shown in FIG. 1 and provides the disease warning information by causing the terminal 41 such as a personal computer, a tablet, or a smartphone used by each person 40 to download the document file. It is good as well.
  • the disease caution information provision support server 10 may print the disease caution information on a paper medium 32 or output a document file and provide it to each person.
  • the disease caution information provision support system is illustrated in FIG. 4A, from the health condition information extraction (STEP 10) by the terminal 30 to the preparation process (STEP 40) by the disease caution information provision support server 10, as shown in FIG. 4B.
  • the process from the health state prediction process (STEP 50) by the disease attention information provision support server 10 to the output of the disease attention information (STEP 120) by the terminal 30 is executed.
  • the disease attention information provision support server 10 or the terminal 30 performs processing on the health condition information or the item of the health condition information
  • the health condition information or the health condition to be processed by the disease attention information provision support server 10 or the terminal 30 The item of information is a part or all of the item.
  • the process shown in FIG. 4A is a series of processes for accumulating the health condition information of each person 40 and preparing for the execution of the series of processes shown in FIG. 4B.
  • the process shown in FIG. It is a series of processes for generating and transmitting attention information.
  • the terminal 30 and the disease attention information provision support server 10 may perform the series of processes shown in FIG. 4A and the series of processes shown in FIG. 4B all at once as a series of processes, or at different frequencies. You may go.
  • the terminal 30 takes out health status information including information on the health status of one or more persons from the storage device (STEP 10).
  • the retrieval may be performed by storing health state information in an external storage medium 31 such as a CD-ROM, DVD-ROM, or USB memory, or by specifying a transmission destination via the information communication network 20. It may be done by sending health status information.
  • the disease caution information provision support server 10 takes in the health condition information taken out by the terminal 30 (STEP 20) and stores it in the health condition storage unit 111 (STEP 30).
  • the disease attention information provision support server 10 executes a preparation process (STEP 40), and makes preparations necessary for predicting the health condition information and quantifying the possibility of developing the disease. Details of the preparation process (STEP 40) will be described later.
  • the terminal 30 and the disease attention information provision support server 10 may perform the series of processes shown in FIG. 4A all at once as a series of processes. Alternatively, for example, only the STEPs 10 to 30 are performed each time, and the data is stored. STEP 40 may be executed every time a certain amount is accumulated or at regular intervals.
  • the prediction unit 102 performs a health state prediction process (STEP 50).
  • the onset probability numerical unit 101 executes the onset probability digitization process (STEP 60) to digitize the onset probability of the disease of each person 40, and the onset probability numerical information is stored as the onset probability numerical information storage unit. It memorize
  • the attention information generation unit 103 executes attention information generation processing (STEP 80) to generate disease attention information including information indicating the numerical value of at least one disease onset of each person 40, and the attention information transmission unit 104 transmits the disease attention information including onset probability numerical value information to the terminal 30 (STEP 90).
  • the information indicating the numerical value of the probability of developing the disease is not only information clearly indicating the numerical value of the probability of developing the disease, but also information that roughly (schematically) indicates the numerical value of the probability of developing the disease. Also good.
  • the information indicating the numerical value of the probability of developing the disease includes not only the numerical value of the probability of developing the disease itself but also information such as a graph indicating the numerical value of the probability of developing the disease and the color indicating the numerical value of the probability of developing the disease. It may be.
  • the terminal 30 receives (step 100), stores (STEP 110), outputs (STEP 120), and outputs (STEP 120) processing of the disease attention information including the numerical value information about the possibility of onset transmitted from the disease attention information provision support server 10. .
  • the terminal 30 and the disease attention information provision support server 10 may perform the series of processes shown in FIG. 4B all at once as a series of processes, or, for example, perform steps 50 to 110 every time, STEP 120 may be executed every time a certain amount is accumulated or at regular intervals.
  • the preparation process includes a change correlation item extraction process (STEP 401), a second time point information acquisition model preparation process (STEP 402), an onset possibility correlation item extraction process (STEP 403), and an onset possibility calculation model preparation process (STEP 404). Composed.
  • the disease caution information provision support server 10 may perform all of these processes at once as a series of processes, or may perform each of them at a different frequency.
  • the change correlation item extraction process (STEP 401) is a process of extracting items of health state information correlated with the magnitude of change in the value of each item of health state information by analyzing the health state information.
  • the processing content is as shown in FIG. 6A and is performed for each item (loop L1) included in the health condition information.
  • the prediction unit 102 calculates the value of each item of the latest health state information of a plurality of persons stored in the health state storage unit 111, the value of each item of health state information one year ago, and two years ago.
  • the value of each item of the health status information is acquired (STEP 4011).
  • the prediction unit 102 calculates the difference between the value of each item of the acquired latest health condition information and the value of each item of the health condition information one year ago for the plurality of persons (STEP 4012).
  • the value of the latest health condition information of a person is weight 89, abdominal circumference 80, total cholesterol 150, blood sugar level 125, the value of one year ago is weight 86, abdominal circumference 78, If the total cholesterol is 155 and the blood glucose level is 124, the difference between the value of the latest health information and the value of the health information one year ago is +3 for body weight, +2 for waist circumference, -5 for total cholesterol, blood glucose The value is +1.
  • the prediction unit 102 calculates the difference between the value of each item of the acquired health condition information one year ago and the value of each item of the health condition information two years ago for the plurality of persons (STEP 4013).
  • the prediction unit 102 calculates the difference between the acquired value of the health condition information one year ago and the value of the health condition information two years ago by the same calculation as STEP 4012.
  • the prediction unit 102 selects each item of the health state information with respect to the magnitude of the change in the value of the item for the health state information that is the extraction target of the item of the health state information correlated with the magnitude of the change in the value. Is calculated (STEP 4014).
  • the prediction unit 102 calculates the degree of correlation, for example, by linear regression analysis.
  • the prediction unit 102 calculates the health status information of one year ago that is the magnitude of the change in the circumference of the subject person.
  • the difference in abdominal circumference between the latest health status is the objective variable, and the difference in the abdominal circumference between the health status information of the person two years ago and the health status of the previous year,
  • the coefficient of each item is obtained by linear regression analysis using a model in which the value of each item contained in the abdominal circumference value and the value of each item included in the health status information of the person one year ago is an explanatory variable.
  • Items with a large absolute value of the coefficient of each health condition information item obtained by the above analysis are items of health condition information having a high degree of correlation with the magnitude of change in the abdominal circumference value (for example, smoking habits, exercise habits, Drinking habits, age and gender).
  • the prediction unit 102 extracts a certain number of items having a high degree of correlation from the items of health state information (STEP 4015). For example, the prediction unit 102 extracts items having a high degree of correlation by a method such as selecting 10 items from the items of the health condition information in the order of the degree of the correlation.
  • the prediction unit 102 correlates the extracted item name with the change in the value of the item of the health condition information, with the item name of the target health condition information as the prediction target, for example, in a format as shown in FIG. 3C.
  • Each item is stored in the change correlation item storage unit 114 as a certain item (STEP 4016).
  • the prediction unit 102 exits the loop L1 after completing these processes (STEPs 4011 to 4016) for all items of the health condition information, and ends the change correlation item extraction process.
  • the second time point information acquisition model preparation process (Second time point information acquisition model preparation process) is a model in which information including the health state information of an arbitrary person at an arbitrary time point is input and information indicating the health state of the person at a later time point is output. This is a process of deriving the constants used for the second time point information acquisition model and the coefficient of each item of the health condition information, and generating or updating the model.
  • the processing content is as shown in FIG. 6B, and is performed for each item of health status information (loop L2).
  • the prediction unit 102 calculates the value of each item of the latest health state information of a plurality of persons stored in the health state storage unit 111, the value of each item of health state information one year ago, and two years ago.
  • the value of each item of the health condition information is acquired (STEP 4021).
  • the prediction unit 102 calculates the difference between the value of each item of the acquired latest health condition information and the value of each item of the health condition information one year ago for the plurality of persons (STEP 4022).
  • the prediction unit 102 calculates a difference between the value of each item of the acquired health condition information one year ago and the value of each item of the health condition information two years ago for the plurality of persons (STEP 4023).
  • the prediction unit 102 refers to the information stored in the change correlation item storage unit 114, and acquires the item name of the health state information correlated with the magnitude of the change in the value of the target item (STEP 4024).
  • the prediction unit 102 determines that the item name of the health condition information stored in the change correlation item storage unit 114 as the item correlated with the change in the value of the health condition information item is smoking. Recognize habits, exercise habits, drinking habits, age and gender.
  • the prediction unit 102 derives constants used for the second time point information acquisition model of the target health condition information item and coefficients of each item of the health condition information (STEP 4025).
  • This model is the same type as the model in the change correlation item extraction process (STEP 401), and is, for example, a linear regression model.
  • the prediction unit 102 determines whether the health status information of one year ago that is the magnitude of the change in the abdominal circumference of the target person and the latest health status.
  • the difference in abdominal circumference between the health status information of the person 2 years ago and the health status of the previous year, the value of the abdominal circumference of the person 1 year ago, and the A constant and a coefficient of each item of a model in which the value of the item correlated with the magnitude of the change in the abdominal circumference among items included in the health condition information of one year ago is obtained by linear regression analysis.
  • the prediction unit 102 generates the second time point information acquisition model of the item of the health condition information of the target by applying the constant obtained by such processing and the coefficient of each item to the equation of the linear regression analysis (STEP 4026).
  • the prediction unit 102 stores the generated second time point information acquisition model of each item of the health condition in the prediction model storage unit 115 (STEP 4027).
  • the prediction unit 102 When the second time point information acquisition model of the item in the health state is already stored in the prediction model storage unit 115, the prediction unit 102 overwrites and stores the generated second time point information acquisition model. Then, the second time point information acquisition model of the item of the health condition is updated.
  • the prediction unit 102 exits the loop L2 after completing these processes (STEPs 4021 to 4027) for all items of the health condition information, and ends the second time point information acquisition model preparation process.
  • the onset probability correlation item extraction process (STEP 403) is a process of extracting items having a correlation with the onset possibility of each disease by analyzing health condition information.
  • the processing content is as shown in FIG. 7A and is performed for each disease (loop L3) stored in the onset possibility correlation item storage unit 112.
  • the onset probability quantification unit 101 extracts the latest health state information of a person who can know the onset of the target disease from the health state information of a plurality of people stored in the health state storage unit 111 (STEP 4031). ).
  • the onset probability quantification unit 101 determines whether or not the onset of the disease is known with reference to, for example, the name of the disease under treatment included in the health condition information.
  • the onset probability digitization unit 101 calculates the degree of correlation of each item of the health status information with respect to the onset of the target disease (STEP 4032).
  • the onset probability digitization unit 101 calculates the degree of correlation by, for example, logistic regression analysis.
  • the onset probability quantification unit 101 uses the onset probability of diabetes of the target person as an objective variable and is included in the health status information of the person
  • the partial regression coefficient of each item of the model using the value of each item as an explanatory variable is obtained by logistic regression analysis.
  • the items with a large absolute value of the partial regression coefficient of each item of the health condition information obtained by the above analysis are items having a high degree of correlation with the onset of diabetes.
  • the onset probability digitization unit 101 extracts a certain number of items having a high degree of correlation from the items of the health condition information (STEP 4033).
  • the onset probability quantification unit 101 extracts, for example, a method such as selecting 10 items in descending order of the degree of correlation.
  • the onset probability quantification unit 101 uses the extracted disease name as an item correlated with the onset probability of each disease in the format as shown in FIG. 3B, for example. Each is stored in the onset likelihood correlation item storage unit 112 (STEP 4034).
  • onset probability quantification unit 101 finishes these processes (STEPs 4031 to 4034) for all the diseases stored in the onset probability correlation item storage unit 112, it exits the loop L3 and extracts the onset probability correlation item. End the process.
  • the onset probability calculation model preparation process (STEP 404) is an onset probability calculation model which is a model that receives the health status information of an arbitrary person at an arbitrary time point and outputs a numerical value of the onset probability for each person's disease. This is a process for deriving coefficients for each item of constants and health condition information to be used.
  • the processing content is as shown in FIG. 7B and is performed for each disease (loop L4) stored in the onset possibility correlation item storage unit 112.
  • the onset probability quantification unit 101 refers to the onset probability correlation item storage unit 112 and acquires the item name of the health condition information correlated with the onset probability of the target disease (STEP 4041).
  • the onset probability quantification unit 101 stores the blood sugar level, waist circumference, drinking habits, HbA1c, exercise habits, smoking, as items that are correlated with the onset possibility. Recognize habits, height, weight, age and gender.
  • the onset probability quantification unit 101 includes the possibility of onset of the target disease acquired in STEP 4041 among the items of the latest health state information of a plurality of persons stored in the health state storage unit 111.
  • the value of each item of the health condition information correlated with is extracted (STEP 4042).
  • the onset probability quantification unit 101 derives constants used in the onset probability calculation model of the target disease and coefficients of each item of the health condition information (STEP 4043).
  • This model is the same as the model in the onset possibility correlation item extraction process (STEP 403), and is, for example, a logistic regression model.
  • the onset possibility quantification unit 101 uses the target person's diabetes onset probability as an objective variable and is extracted in STEP 4042 A constant and a partial regression coefficient of each item of a model having the value of an item correlated with the possibility of developing the target disease as an explanatory variable are obtained by logistic regression analysis.
  • the onset probability quantification unit 101 generates a model for calculating the onset probability of the target disease by applying the constants obtained by such processing and the partial regression coefficient of each item to the logistic regression analysis formula (STEP 4044). .
  • the onset probability quantification unit 101 stores the generated or updated onset probability calculation model of each disease in the calculation model storage unit 113 (STEP 4045).
  • the onset probability digitizing unit 101 overwrites and stores the generated onset probability calculation model. Update the onset probability calculation model for the item of health status.
  • the onset probability quantification unit 101 exits the loop L4 when all the diseases stored in the onset probability correlation item storage unit 112 have been completed (STEP 4041 to STEP 4045), and performs the onset probability quantification model processing. finish.
  • the health state prediction process (STEP 50) is a process for acquiring information indicating the predicted health state of each person 40 at the second time point. Details of this processing will be described with reference to FIGS.
  • the processing content is as shown in FIG. 8 and is performed for each ID (loop L5) included in the health condition information stored in the health condition storage unit 111. Further, STEPs 503 to 506 are performed for each item of health condition information (loop L6).
  • the prediction unit 102 acquires the health state information J1 and the latest health state information J2 of the target ID one year ago from the health state storage unit 111 (STEP 501, FIG. 9).
  • the prediction unit 102 calculates a difference J11 between the value of each item of the acquired latest health condition information and the value of each item of the health condition information one year ago for the ID (STEP 502, FIG. 9).
  • the prediction unit 102 calculates the difference J11 between the value of the latest health condition information and the value of the health condition information one year ago. It may be included in the processing target in the subsequent processing, or the difference J11 between the value of the latest health state information and the health state information one year ago is not calculated, and is not included in the processing target in the subsequent processing. It is good.
  • the prediction unit 102 sets the value of each item such as 1 for habit and 0 for no habit, for example, if it is a smoking habit. Replace with numeric values.
  • the prediction unit 102 refers to the change correlation item storage unit 114, and acquires items of health state information correlated with the magnitude of change of the target health state information item (STEP 503).
  • the prediction unit 102 extracts an item J13 having a correlation with the magnitude of change in the item of the target health condition information from the latest health condition information of the target ID for the item of the target health condition information (STEP 504). , FIG. 9).
  • the prediction unit 102 extracts the values of these items from the latest health condition information of the ID.
  • the prediction unit 102 refers to the prediction model storage unit 115 and acquires the second time point information acquisition model F1 of the item of the target health condition information (STEP 505, FIG. 9).
  • the prediction unit 102 correlates with the difference J11 between the latest value of the target item of the target ID and the value one year ago, the latest value J12 of the target item of the target ID, and the target item. Based on the value of each item J13, the value at the second time point of the target item of the target ID is calculated (STEP 506, FIG. 9).
  • the prediction unit 102 adds the difference J11 between the latest value of the target item and the value one year ago, the latest value J12, and the value of each item J13 correlated with the target item to the model F1 acquired in STEP 505. By inputting, a predicted value V1 of a difference between the latest value of the item of the health condition information of the target and the value at the second time point is calculated (FIG. 9).
  • the prediction unit 102 adds the obtained V1 and the latest value J12 of the item to be predicted, and obtains the predicted value V2 of the item to be predicted one year later as the second time point (FIG. 9). .
  • the prediction unit 102 may obtain, for example, a 95% confidence interval value of the predicted value V1 of the difference between the latest value of the target health condition information item and the value at the second time point.
  • the prediction unit 102 has a certain range as the predicted value V1 of the difference between the latest value of the item of the target health condition information and the value at the second time point and the predicted value V2 of the target item at the second time point.
  • the value can be acquired (FIG. 9).
  • the prediction unit 102 stores the value at the second time point of the target item of the target ID calculated in STEP 506 in the predicted value information storage unit 116 (STEP 507).
  • the prediction unit 102 finishes the processing of STEPs 503 to 507 for all items of the health condition information, it exits the loop L6, and STEPs 501 to 507 for all IDs included in the health condition information stored in the health condition storage unit 111.
  • the process exits the loop L5 and ends the health state prediction process.
  • the onset possibility quantification process (STEP 60) is a process for quantifying the latest onset and second onset illness of each person 40. Details of this processing will be described with reference to FIGS.
  • the onset probability digitization unit 101 repeatedly quantifies the onset probability for each disease of each person 40 by the onset probability digitization process at an arbitrary frequency.
  • the processing content is as shown in FIG. 10 and is performed for each ID (loop L7) included in the health condition information stored in the health condition storage unit 111. Further, STEPs 602 to 609 are performed for each disease (loop L8) stored in the onset possibility correlation item storage unit 112.
  • the symptom probability quantification unit 101 stores the latest health state information of the target ID stored in the health state storage unit 111 and the second time point of the target ID stored in the predicted value information storage unit 116.
  • the onset probability numerical value information of the subject ID stored in the health state information and onset probability numerical information storage unit 117 one year ago is acquired (STEP 601).
  • the onset probability digitizing unit 101 refers to the onset probability correlation item storage unit, and acquires items of health state information correlated with the onset possibility of the target disease (STEP 602).
  • the onset probability digitizing unit 101 extracts the value of the item correlated with the onset probability for each disease from the latest health state information of the target ID (STEP 603).
  • the unit 101 extracts the values of these items from the latest health state information J3 of the target ID (FIG. 11).
  • the onset probability quantification unit 101 refers to the calculation model storage unit 113 and acquires the onset probability calculation model F2 of diabetes (STEP 604, FIG. 11).
  • the onset possibility quantification unit 101 numerically calculates the latest onset possibility of the target disease of the target ID based on the value of each item J31 correlated with the onset possibility of diabetes extracted in STEP 603. (STEP 605, FIG. 11).
  • the onset possibility quantification unit 101 inputs the value of each item J31 correlated with the onset possibility of diabetes extracted in STEP 603 to the model F2 acquired in STEP 604, and the latest onset possibility of the target disease.
  • the numerical value V3 is calculated (FIG. 11).
  • the onset probability quantification unit 101 calculates the onset probability calculation model for dyslipidemia based on the item J32 correlated with the onset possibility of dyslipidemia.
  • the latest possibility of onset is quantified using F3 (FIG. 11).
  • the onset probability digitization unit 101 extracts the value of the item correlated with the onset possibility of the target disease from the information indicating the predicted health state at the second time point of the target ID (STEP 606).
  • the possibility digitizing unit 101 extracts the values of these items from the health state information J4 at the second time point of the target ID acquired in STEP 601 (FIG. 12).
  • the prediction unit 102 when the prediction unit 102 does not include items that are not measured with numerical values such as exercise habits and smoking habits, the likelihood of occurrence quantification unit 101 These pieces of information are extracted from the latest health condition information J3 of the target ID (FIG. 12).
  • the onset possibility quantification unit 101 quantifies the onset possibility at the second time point of the target disease of the target ID based on the item J41 correlated with the extracted onset possibility of diabetes (STEP 607). ).
  • the onset possibility calculation model F2 acquired in STEP 604 is used again.
  • each item J41 correlated with the possibility of the onset of diabetes extracted in STEP 605 is input to the model, and the numerical value V4 of the onset probability at the second time point of the target disease of the target ID is calculated (FIG. 12).
  • STEP 506 of the health state prediction process when the prediction unit 102 acquires a value with a certain width as the predicted value V2 at the second time point of the target item and stores it in the predicted value information storage unit 116.
  • the onset probability digitization unit 101 can calculate a wide value as the onset probability value V4 at the second time point of the target disease.
  • the onset probability quantification unit 101 acquires, for example, the maximum value, the median value, and the minimum value of each item J41 correlated with the onset probability of the disease, and the combination of only the maximum value, the minimum value Only the combination, the maximum value, the median, and any combination of the minimum value, etc., are input to the onset probability calculation model F2 to calculate the numerical value V4 of the onset probability at the second time point of the target disease, Among the obtained calculation result values, the maximum value is the maximum value of the disease onset probability value V4, and the obtained calculation result value is the minimum value of the disease onset probability value V4. Calculated as a value (FIG. 12).
  • the onset probability quantification unit 101 calculates the onset probability calculation model for dyslipidemia based on the item J42 correlated with the onset possibility of dyslipidemia.
  • the latest possibility of onset is quantified using F3 (FIG. 12).
  • the onset probability digitization unit 101 sets the type of onset possibility according to the latest ID of the target ID, the second time point, and the onset probability of the disease one year ago (STEP 608).
  • the onset probability digitizing unit 101 determines and sets the type of onset possibility by referring to the conversion table as shown in FIG. 14A stored in the conversion table storage unit 119 in order from the top, for example.
  • the onset probability digitizing unit 101 sets the type of the onset probability to 4.
  • the numerical value of the probability of developing a certain disease does not correspond to the types 4 and 3 of the probability of developing, and the numerical value of the probability of developing is 100 or less, but a disease that has increased by 50 or more compared to the previous time is 1 If there is more than one, the onset probability quantification unit 101 sets the type of onset possibility to 2.
  • the onset probability quantification unit 101 determines whether there is one or more diseases whose onset probability has increased by 50 or more compared to the previous time, for example, for each disease of the target ID calculated in STEP 605 Compare the latest numerical value of the likelihood of onset and the value of the numerical value of the probability of onset one year before the disease with the ID acquired in STEP601, and one or more diseases whose numerical value of the probability of onset has increased by 50 or more Depending on whether or not there is.
  • the onset probability quantification unit 101 does not correspond to any of the 4 to 2 types of onset possibility of the onset possibility of a certain disease, but the onset possibility value at the second time point is 100 or more.
  • the type of onset possibility is set to 1.
  • the onset probability digitizing unit 101 stores the latest onset probability numerical information on the target ID and the type of onset possibility in the onset probability numerical information storage unit 117, and the disease at the second time point of the target ID. Are stored in the predicted value information storage unit 116 (STEP 609).
  • the onset probability quantification unit 101 exits the loop L8 when all the diseases stored in the onset probability correlation item storage unit 112 have been processed in STEPs 602 to 609, and is stored in the health state storage unit 111.
  • the processing of STEPs 601 to 609 is completed for all the IDs included in the health status information, the onset probability quantification processing is terminated.
  • the attention information generation process (STEP 80) is a process for generating disease attention information including information on the possibility of the onset of the disease according to the numerical value of the onset of the disease of each person 40.
  • the processing content is as shown in FIG. 13 and is performed for each ID (loop L9) included in the health condition information stored in the health condition storage unit 111.
  • the generation of the disease attention information by the attention information generation unit 103 is repeatedly performed at an arbitrary frequency.
  • the attention information generation unit 103 acquires the latest type of possibility of onset of the target ID (STEP 801), and recognizes the generation frequency of the disease attention information according to the type of the possibility of onset (STEP 802).
  • the attention information generation unit 103 refers to the conversion table as shown in FIG. 14B stored in the conversion table storage unit 119, for example, and stores the disease attention information according to the type of the latest disease onset possibility of the target ID. Recognize the generation frequency.
  • the attention information generation unit 103 sets the generation frequency of the disease attention information of the target ID as monthly when the maximum value of the type of the possibility of developing the disease of the target ID is 4.
  • the attention information generation unit 103 determines whether or not a time equal to or greater than the generation frequency of the disease attention information has elapsed since the disease attention information was previously generated for the target ID (STEP 803).
  • the attention information generation unit 103 refers to the attention information last date information as shown in FIG. 3F for the determination, and performs the attention information last date of each target ID and the attention information concerned processing. This is done by comparing the date and determining whether or not a time equal to or greater than the generation frequency has elapsed.
  • the attention information generation unit 103 determines that the time more than the generation frequency has passed if half a year has passed since the last date of the attention information about the ID. If not, it is determined that the time exceeding the generation frequency has not elapsed.
  • Step 803 NO
  • the attention information generation unit 103 does not execute STEPs 804 to 805, After transferring the person to be processed to the next person, the processing after STEP 801 is repeated.
  • the attention information generation unit 103 creates the disease attention information.
  • the latest health condition information of the target ID, the health condition information one year ago, and the health condition information at the second time point are acquired (STEP 804).
  • the attention information generation unit 103 generates disease attention information according to the numerical value of the possibility of onset of one or both of the latest disease of the target ID and the second time point (STEP 805).
  • the attention information generation unit 103 generates the improvement advice information 322 with reference to the conversion table as illustrated in FIG. 14C stored in the conversion table storage unit 119, for example.
  • the attention information generation unit 103 may “relieve obesity, reduce the amount of meals, excessive intake of animal fats and carbohydrates (especially soft drinks)”.
  • the improvement advice information 322 is generated by acquiring the improvement advice corresponding to the type of the disease and the possibility of onset such as “Reserve ...”.
  • the attention information generation unit 103 may be configured to generate the improvement advice information only for a disease for which the numerical value of the possibility of developing the disease is a certain value or more.
  • the attention information generation unit 103 generates complication information 323 with reference to a conversion table as illustrated in FIG. 14D stored in the conversion table storage unit 119, for example.
  • the attention information generation unit 103 obtains information on complications such as “myocardial infarction, stroke, renal failure” for those who have or may develop diabetes. Information 323 is generated.
  • the attention information generation unit 103 may be configured to arrange the graph 324 representing the numerical value of the onset probability for each disease in descending order of the numerical value of the onset probability for each disease.
  • the attention information generation unit 103 creates a group composed of all or part of a plurality of people for each ID having the same gender and the same age, for example, and ranks of the onset probability for each disease of each ID in the group Is calculated to generate information 325 relating to the ranking of the onset probability of each person 40 for each disease.
  • the attention information generation unit 103 may calculate the actual rank within the group and may represent, for example, the 50th place in 1975, or the rank within the group may be replaced with the rank of the 100 person. If there are numbers after the decimal point, they may be rounded off and expressed as 3rd place out of 100 people.
  • the illness caution information provision support server 10 captures information such as the work name of each person 40 from the terminal 30, the business type of the work, and the occupation type, and the caution information generation unit 103 uses a plurality of people based on the information. It is good also as creating the group comprised by all or one part.
  • the attention information generation unit 103 includes improvement advice information 322, information 323 relating to complications, information 325 relating to the ranking of the onset probability for each disease of each ID in a group composed of all or part of a plurality of persons, and the like.
  • the information may be generated based on not only the latest numerical value of the onset possibility for each disease of each ID but also the numerical value of the onset possibility at the second time point for each disease.
  • the attention information generation unit 103 stores the generation date for each ID of the disease attention information in the history storage unit 118 (STEP 806).
  • the attention information generation unit 103 exits the loop L9 when these processes (STEPs 801 to 806) are completed for all IDs included in the health state information stored in the health state storage unit 111, and ends the attention information generation process. To do.
  • the model or prediction unit 102 used when the onset probability quantification unit 101 quantifies the onset probability of each disease indicates the predicted health state of the target person at a second time point after the first time point.
  • an identification machine such as deep learning or a support vector machine may be used.
  • DESCRIPTION OF SYMBOLS 10 ... Disease attention information provision support server, 100 ... Server control part, 101 ... Probability digitization part, 102 ... Prediction part, 103 ... Caution information generation part, 104 ... Caution information transmission part, 110 ... Server memory part, 111 ... health condition storage unit, 112 ... onset possibility correlation item storage unit, 113 ... calculation model storage unit, 114 ... change correlation item storage unit, 115 ... prediction model storage unit, 116 ... prediction value information storage unit, 117 ... onset possibility Sex numerical value information storage unit, 118 ... history storage unit, 119 ... conversion table storage unit, 20 ... information communication network, 30 ... terminal, 40 ... each person.

Landscapes

  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Data Mining & Analysis (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Biomedical Technology (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

Provided is a disease-warning-information providing support system that enables a person who has developed a disease or who has a possibility of developing a disease to adequately recognize the need for improvement relating thereto, thereby promoting visits to medical institutions and improvements in health status through the visits. A disease-warning-information providing support system according to the present invention is provided with: a disease-developing-possibility quantifying unit that quantifies per-disease disease developing possibilities of each person on the basis of health status information; and a warning-information generating unit that generates disease warning information including information indicating numerical values concerning the possibilities of developing diseases.

Description

疾病注意情報提供支援システムDisease attention information provision support system
　本発明は、疾病注意情報提供支援システムに関する。 The present invention relates to a disease attention information provision support system.
　従来、健康保険組合の被保険者が生活習慣病などの疾病を患っていることが判明した場合や、医療機関で受診している場合に、当該被保険者の医療機関での受診の要否を判別して医師等に知らせる医療受診支援システムが知られている（例えば、特許文献１参照）。 Previously, if it was found that an insured person of a health insurance association was suffering from a disease such as a lifestyle-related disease, or if he / she was receiving a medical examination at a medical institution, whether or not the insured should be examined at a medical institution A medical consultation support system that discriminates and informs a doctor or the like is known (see, for example, Patent Document 1).
　特許文献１の医療受診支援システムでは、被保険者の病気の進行と診療報酬の増加を抑制するために、被保険者のレセプトに記載された内容と健康診断の結果とから、被保険者について、受診病名と当該受診病名に関連する健康診断の検査結果データとを抽出し、抽出された検査結果データを当該受診病名の医療ガイドラインと比較し、当該受診病名の病気の進行度を表す指標を演算し、演算結果に基づいて当該被保険者の医療機関での受診の要否を判別する。この判別の結果は、医師等が被保険者に対して医療機関での受診継続を勧告するなどの指導を行う際に用いることができる。 In the medical examination support system of Patent Document 1, in order to suppress the progression of the insured's illness and the increase in medical fee, the contents described in the insured's receipt and the result of the medical examination are used to determine the insured person. , Extract the name of the diagnosis and the test result data of the health check related to the name of the diagnosis, compare the extracted test result data with the medical guidelines for the name of the diagnosis, and provide an index indicating the degree of disease progression of the diagnosis The calculation is performed, and it is determined whether or not the insured needs to have a medical examination at a medical institution based on the calculation result. The result of this determination can be used when a doctor or the like gives instructions to the insured such as recommending continued medical care at a medical institution.
特開２００４－１６４１７３号JP 2004-164173 A
　医療の効率化のためには、この種の情報を各被保険者に対してそのまま通知できることが好ましい。 In order to improve medical efficiency, it is preferable that this type of information can be notified to each insured person as it is.
　しかしながら、特許文献１に開示されているシステムで得られる情報は、傷病名、血糖値や総コレステロール値などの検査結果の数値及び、当該数値に基づく「受診継続」、などの専門知識を有する者が閲覧したことを前提とした情報である。 However, the information obtained by the system disclosed in Patent Document 1 is a person who has specialized knowledge such as the name of injury and illness, numerical values of test results such as blood glucose level and total cholesterol level, and “continuation of consultation” based on the numerical values. It is information on the assumption that has been viewed.
　このため、特許文献１に開示されているシステムで得られた情報を専門知識を有しない被保険者が閲覧しても、被保険者はどの程度自己の病状が深刻なのかをにわかには理解することができない。 For this reason, even if an insured who does not have specialized knowledge views the information obtained by the system disclosed in Patent Document 1, the insured understands how serious his / her medical condition is. I can't.
　したがって、特許文献１に開示されているシステムで得られた情報をそのまま被保険者に閲覧させても、被保険者にその病状の深刻さを理解させることができず、被保険者が、医療機関での受診を怠ったり、自己の健康管理を怠ったりするなどし、場合によっては健康状態が悪化してしまうおそれがある。 Therefore, even if the insured person browses the information obtained by the system disclosed in Patent Document 1 as it is, the insured person cannot understand the seriousness of the medical condition. There is a risk that health may deteriorate in some cases, such as neglecting medical examinations at the institution or neglecting self-care.
　本発明の目的は、かかる従来技術の課題に鑑み、疾病を発症している又は発症する可能性を有する者にその改善の必要性を十分に認識させ、医療機関での受診や受診を通した健康状態の改善を促進することができる疾病注意情報提供支援システムを提供することにある。 The purpose of the present invention is to allow a person who has developed or has a possibility of developing a disease to fully recognize the necessity of improvement in view of the problems of the prior art, and through medical examinations and medical examinations. It is to provide a disease attention information providing support system that can promote improvement of health condition.
　本発明の疾病注意情報提供支援システムは、
　１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶する健康状態記憶部と、
　記憶した前記健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する発症可能性数値化部と、
　前記発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部と、
　任意の時点における任意の人の健康状態情報を含む情報を入力として当該時点より後の時点における該人の健康状態を示す情報を出力とするモデルに、第１時点における対象の人の健康状態情報を含む情報を入力し、該モデルから出力された情報により、該第１時点より後の第２時点における該対象の人の予測される健康状態を示す情報を取得する予測部を備え、
　前記発症可能性数値化部は、前記予測部が取得した情報に基づいて各人の疾病ごとの前記第２時点における発症可能性を数値化し、
　前記注意情報生成部は、疾病ごとの前記第２時点における発症可能性の数値に基づいて前記疾病注意情報を生成するように構成されていることを特徴とする。 The disease attention information provision support system of the present invention is
A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons;
An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information;
Attention information generation unit for generating disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or a plurality of diseases digitized by the onset probability digitization unit;
Information on the health status of the subject at the first time is input to the model including the information including the health status information of any man at any time as an input and information indicating the health status of the man at a time after the time is output. Including a prediction unit that acquires information indicating a predicted health state of the target person at a second time point after the first time point according to the information output from the model,
The onset probability quantification unit quantifies the onset probability at the second time point for each person's disease based on the information acquired by the prediction unit,
The attention information generation unit is configured to generate the disease attention information based on a numerical value of the onset possibility at the second time point for each disease.
　本発明によれば、まず発症可能性数値化部が１又は複数の人の健康状態に関する情報を含む健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する。 According to the present invention, the onset probability quantification unit first quantifies the onset probability for each person's disease based on the health condition information including information on the health condition of one or more persons.
　次に、注意情報生成部が発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する。 Next, the attention information generation unit generates disease attention information including information indicating the numerical value of the onset probability of at least one of the one or more diseases digitized by the onset probability digitization unit.
　このように、本発明によれば、各人の疾病ごとの発症可能性を数値化し、1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報が生成される。 As described above, according to the present invention, the possibility of developing each person's disease is quantified, and disease attention information including information indicating the numerical value of the probability of developing at least one disease among one or more diseases is generated. Is done.
　そのため、各人の疾病ごとの発症可能性を数値化することで、発症可能性が数値として理解できるようになるとともに、1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を与えることが可能になるので、疾病を発症している又は発症する可能性を有する者にその改善の必要性を十分に認識させ、医療機関での受診や受診を通した健康状態の改善を促進することができる。 Therefore, by quantifying the probability of onset for each person's illness, it becomes possible to understand the possibility of onset as a numerical value, and also shows the numerical value of the likelihood of developing at least one disease among one or more diseases Since it is possible to give information on disease cautions including information, make sure that people who have or have the potential to develop a disease are fully aware of the need for improvement and receive medical examinations and medical examinations. Can promote improved health.
　そして、本発明の疾病注意情報提供支援システムは、第１時点における対象の人の健康状態情報を含む情報を用いて、当該第１時点より後の第２時点における当該対象の人の予測される健康状態を示す情報を取得する予測部を備えている。 The disease attention information provision support system of the present invention predicts the target person at a second time point after the first time point using information including the health condition information of the target person at the first time point. A prediction unit is provided for acquiring information indicating a health condition.
　予測部は、任意の時点における任意の人の健康状態情報を含む情報を入力として当該時点より後の時点における当該人の健康状態を示す情報を出力とするモデルを用いて、第１時点より後の第２時点における対象の人の予測される健康状態を示す情報を取得する。 The prediction unit uses information including information on the health status of any person at any time as an input, and outputs information indicating the health status of the person at a time later than that time. Information indicating the predicted health state of the subject person at the second time point is acquired.
　また、発症可能性数値化部は、前記予測部が取得した情報に基づいて各人の疾病ごとの第２時点における発症可能性を数値化する。 Also, the onset probability digitization unit quantifies the onset probability at the second time point for each person's disease based on the information acquired by the prediction unit.
　さらに、注意情報生成部は、疾病ごとの第２時点における発症可能性の数値に基づいて疾病注意情報を生成する。 Further, the attention information generation unit generates disease attention information based on a numerical value of the possibility of onset at the second time point for each disease.
　これにより、ある時点で疾病を発症している又は発症する可能性を有する者だけでなく将来的に疾病を発症する可能性がある者に対する疾病注意情報を生成することができるので、より広範囲の者に対して疾病注意情報を提供することができる。 As a result, it is possible to generate disease caution information not only for those who develop or have the possibility of developing a disease at a certain point of time, but also for those who may develop the disease in the future. Information on illness can be provided to the person.
　本発明の疾病注意情報提供支援システムにおいて、
　前記疾病注意情報は、疾病ごとの発症可能性の改善方法に関する情報である改善アドバイス情報を含み、
　前記注意情報生成部は、疾病の発症可能性の数値が一定値以上である疾病についてのみ前記改善アドバイス情報を生成するように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
The disease attention information includes improvement advice information that is information on a method for improving the possibility of onset for each disease,
It is preferable that the attention information generation unit is configured to generate the improvement advice information only for a disease having a numerical value of a disease onset probability equal to or greater than a certain value.
　本発明によれば、疾病注意情報は、疾病ごとの発症可能性の改善方法に関する情報である改善アドバイス情報を含んでいる。 According to the present invention, the disease attention information includes improvement advice information that is information related to a method for improving the possibility of onset for each disease.
　また、当該改善アドバイス情報は、疾病の発症可能性の数値が一定値以上である疾病についてのみ生成される。 Further, the improvement advice information is generated only for a disease whose numerical value of the probability of developing the disease is a certain value or more.
　これにより、疾病の発症可能性の数値が一定値以上である疾病についてのみ改善アドバイス情報が生成されるので、改善の優先度が高い疾病をより強く印象付けることができる。 Thereby, the improvement advice information is generated only for the disease for which the numerical value of the onset of the disease is a certain value or more, so that the disease having a high priority for improvement can be more strongly impressed.
　本発明の疾病注意情報提供支援システムにおいて、
　前記疾病注意情報は、疾病の発症可能性の数値に応じて各人が発症する可能性のある疾病に起因して発症する可能性のある合併症に関する情報を含むように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
The disease attention information is configured to include information on complications that may occur due to a disease that each person may develop according to a numerical value of the possibility of developing the disease. preferable.
　本発明によれば、疾病注意情報は、各人が発症する可能性のある疾病に起因して発症する可能性のある合併症に関する情報を含んでいるので、疾病を発症している又は発症する可能性を有する者にその改善の必要性を、個別の疾病について情報提供した場合よりも強く印象付けることができる。 According to the present invention, the disease attention information includes information on complications that may develop due to the diseases that each person may develop, so that the disease has or has developed. The potential for improvement can be impressed by those who have the potential more than when information is provided about individual diseases.
　本発明の疾病注意情報提供支援システムにおいて、
　前記疾病注意情報は、各人の疾病ごとの発症可能性の数値を表すグラフを含み、該グラフは疾病ごとの発症可能性の数値が高い順に並べられていることが好ましい。 In the disease attention information provision support system of the present invention,
It is preferable that the disease attention information includes a graph representing a numerical value of the likelihood of onset for each person's disease, and the graph is arranged in descending order of the numerical value of the probability of onset for each disease.
　本発明によれば、疾病注意情報は、各人の疾病ごとの発症可能性の数値を表すグラフを含んでいる。また、当該グラフは、疾病ごとの発症可能性の数値が高い順に並べられている。 According to the present invention, the disease attention information includes a graph representing the numerical value of the onset possibility for each person's disease. Moreover, the graph is arranged in descending order of the numerical value of the onset probability for each disease.
　これにより、いずれの疾病が発症している又は発症する可能性が高いかが一目瞭然となるので、各人にその改善の必要性を明確に認識させることができる。 This makes it clear at a glance which disease has or is likely to develop, so that each person can clearly recognize the need for improvement.
　本発明の疾病注意情報提供支援システムにおいて、
　前記疾病注意情報は、前記複数の人の全部又は一部により構成されるグループにおける各人の疾病ごとの発症可能性の順位に関する情報を含むように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
It is preferable that the disease attention information is configured to include information related to the ranking of the probability of occurrence of each person's disease in the group formed by all or part of the plurality of persons.
　本発明によれば、疾病注意情報は、当該複数の人の全部又は一部により構成されるグループにおける各人の疾病ごとの発症可能性の順位に関する情報を含んでいる。 According to the present invention, the disease attention information includes information related to the ranking of the onset probability of each person's disease in a group composed of all or part of the plurality of persons.
　そのため、各疾病の発症可能性の高さについて、他者と比較した相対的な順位を知らせることができるので、各疾病の発症可能性を数値のみにより知らせた場合よりも、疾病を発症している又は発症する可能性を有する者にその改善の必要性の高さを簡潔に認識させることができる。 Therefore, because it is possible to inform the relative rank of each disease compared to others about the high probability of onset of each disease, it is possible to develop the disease rather than informing only the possibility of each disease by numerical values. A person who has or has a possibility of developing symptoms can be made aware of the high need for improvement.
　本発明の疾病注意情報提供支援システムにおいて、
　前記発症可能性数値化部は、前記健康状態情報から疾病ごとの発症可能性と相関のある健康状態情報を抽出し、相関のある健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化するように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
The onset probability quantification unit extracts health state information correlated with the onset probability for each disease from the health state information, and determines the onset probability for each person's disease based on the correlated health state information. It is preferable to be configured to be numerical.
　本発明によれば、健康状態情報の中から疾病ごとの発症可能性と相関のある健康状態情報を抽出したうえで、それらの情報に基づいて疾病ごとの発症可能性を数値化するので、健康状態情報のすべての項目を考慮するよりも効率的に数値化処理を行うことができる。 According to the present invention, after extracting the health condition information correlated with the possibility of onset for each disease from the health condition information, the possibility of onset for each disease is quantified based on the information. Numerical processing can be performed more efficiently than considering all items of state information.
　本発明の疾病注意情報提供支援システムにおいて、
　前記予測部は、複数の人のそれぞれのある時点における健康状態情報を含む情報と、該複数の人のそれぞれの該時点の健康状態情報と該複数の人のそれぞれの該時点より前の時点の健康状態情報との間における変化の大きさを含む情報とを用いて、前記モデルを生成又は更新するように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
The prediction unit includes information including health state information at each time point of the plurality of people, health state information at each time point of the plurality of people, and time points before the respective time points of the plurality of people. It is preferable that the model is generated or updated using information including the magnitude of change between the health status information.
　本発明によれば、予測部は、複数の人のそれぞれのある時点における健康状態情報を含む情報と、該複数の人のそれぞれの該時点の健康状態情報と該複数の人のそれぞれの該時点より前の時点の健康状態情報との間における変化の大きさを含む情報と用いて、第１時点より後の第２時点における対象の人の予測される健康状態を示す情報を取得するために用いるモデルを生成又は更新する。 According to the present invention, the prediction unit includes information including health state information at a certain time point of each of the plurality of persons, health state information of each of the plurality of persons, and each time point of the plurality of persons. In order to obtain information indicating the predicted health state of the target person at a second time point after the first time point using information including the magnitude of change between the health state information at an earlier time point Generate or update the model to use.
　これにより、予測部は複数の人の実際の健康状態情報の変化の大きさを含む情報を用いて当該モデルを更新するので、第２時点における対象の人の予測される健康状態を示す情報の精度を高めることができる。 Thereby, since the prediction unit updates the model using information including the magnitude of the change in the actual health status information of a plurality of people, the information indicating the predicted health status of the target person at the second time point Accuracy can be increased.
　本発明の疾病注意情報提供支援システムにおいて、
　前記発症可能性数値化部は、各人の疾病ごとの発症可能性の数値化を繰り返し行い、前記注意情報生成部は、少なくとも該数値が一定以上変化した人向けの疾病注意情報を生成するように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
The onset probability quantification unit repeatedly quantifies the onset probability of each person's disease, and the attention information generation unit generates disease attention information for a person whose value has changed at least a certain level. It is preferable that it is comprised.
　本発明によれば、発症可能性数値化部が各人の疾病ごとの発症可能性の数値化を繰り返し行い、注意情報生成部が疾病注意情報を生成するので、一度のみ疾病注意情報を生成する場合に比して、疾病を発症している又は発症する可能性を有する者にその改善の必要性をより強く認識させることができる。 According to the present invention, the onset probability quantification unit repeatedly quantifies the onset probability for each person's disease, and the attention information generation unit generates the disease attention information, so the disease attention information is generated only once. As compared with the case, it is possible to make a person who has developed or has a possibility of developing a disease more strongly aware of the need for improvement.
　また、少なくとも発症可能性の数値が一定以上変化した人向けの疾病注意情報を生成するので、例えば疾病を発症する可能性の値が急激に大きくなった人に対しては、疾病を発症する危険性が急激に高まっている状況をその都度確実に認識させることができる。 In addition, since the disease attention information is generated for people who have at least a certain change in the likelihood of developing the disease, for example, the risk of developing the disease is increased for those who have a sudden increase in the likelihood of developing the disease. It is possible to reliably recognize the situation where the nature is rapidly increasing.
　これにより、医療機関での受診や受診を通した健康状態のより確実な改善を促進することができる。 This makes it possible to promote more reliable improvement of health through medical examinations and medical examinations.
　本発明の疾病注意情報提供支援システムにおいて、
　前記注意情報生成部は、各人の疾病ごとの発症可能性の数値に応じて疾病注意情報を生成する頻度を異なるものとするように構成されていることが好ましい。 In the disease attention information provision support system of the present invention,
It is preferable that the attention information generation unit is configured to vary the frequency of generating the disease attention information according to the numerical value of the possibility of onset for each person's disease.
　本発明によれば、注意情報生成部は、各人の疾病ごとの発症可能性の数値に応じて疾病注意情報を生成する頻度を変える。 According to the present invention, the attention information generation unit changes the frequency of generating the disease attention information according to the numerical value of the possibility of onset for each person's disease.
　これにより、例えば発症可能性の数値が高いため医療機関での早期の受診が望まれる人には頻繁に、発症可能性のそれほど高くない人にはある程度長く間隔をあけて疾病注意情報を生成するなど、個別の人の発症可能性の状況を考慮した重み付けをして、それぞれにとって適切な頻度で効率的かつ効果的に医療機関での受診や受診を通した健康状態の改善を促進することができる。 In this way, for example, because of the high possibility of onset, the disease attention information is generated frequently for people who want to have an early visit at a medical institution, and for those who do not have a high possibility of onset at a certain interval. To improve the health status through medical examinations and consultations at medical institutions efficiently and effectively at an appropriate frequency for each person. it can.
　本発明の疾病注意情報提供支援システムは、
　１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶する健康状態記憶部と、
　記憶した前記健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する発症可能性数値化部と、
　前記発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部とを備え、
　前記疾病注意情報は、該疾病注意情報が提供される対象の人の疾病ごとの発症可能性について、前記複数の人の全部又は一部により構成されるグループにおける順位を示す情報を含むように構成されていることを特徴とする。 The disease attention information provision support system of the present invention is
A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons;
An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information;
An attention information generation unit that generates disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or more diseases digitized by the onset probability digitization unit,
The disease attention information is configured to include information indicating a rank in a group composed of all or a part of the plurality of persons with respect to the possibility of onset for each disease of a person to whom the disease attention information is provided. It is characterized by being.
　本発明によれば、疾病注意情報は、該疾病注意情報が提供される対象の人の疾病ごとの発症可能性について、前記複数の人の全部又は一部により構成されるグループにおける順位を示す情報を含んでいる。 According to the present invention, the disease attention information is information indicating a rank in a group composed of all or a part of the plurality of persons, with respect to the possibility of occurrence of each disease of the target person to whom the disease attention information is provided. Is included.
　疾病ごとの発症可能性の数値を各人に提供するだけでは、各人に病状の深刻さを十分に理解させることができず、各人が、医療機関での受診を怠ったり、自己の健康管理を怠ったりするなどし、場合によっては健康状態が悪化してしまうおそれがある。 Just providing each person with the probability of onset for each disease does not allow each person to fully understand the seriousness of the condition, and each person fails to visit a medical institution or has their own health. In some cases, the health status may deteriorate due to neglect of management.
　そのような課題に対応するため、他の人を含めた疾病ごとの発症可能性の数値を一覧できるようにした統計情報を各人に提供することも考えられるが、当該情報の提供を受ける対象者本人の疾病ごとの発症可能性の具体的な順位が直感的に理解できないのであれば、専門知識を有しない各人は病状の深刻さを十分に理解することができない懸念がある。 In order to respond to such issues, it is possible to provide each person with statistical information that can list the numerical value of the likelihood of developing each disease, including other people. If the specific order of the onset probability for each person's disease cannot be intuitively understood, there is a concern that each person who does not have expertise cannot fully understand the seriousness of the medical condition.
　本構成の疾病注意情報提供支援システムによれば、疾病注意情報は、「疾病注意情報が提供される対象の人」の疾病ごとの発症可能性について、「複数の人の全部又は一部により構成されるグループにおける順位を示す情報」を含むように構成されているので、当該情報の提供を受ける対象者本人が疾病ごとの発症可能性の具体的な順位を直感的に理解できる。 According to the disease attention information provision support system of this configuration, the disease attention information is composed of “all or part of a plurality of people” regarding the possibility of occurrence of each disease of “the person to whom the disease attention information is provided”. Since the target person who receives the information can intuitively understand the specific order of the onset probability for each disease.
　このように、本構成の疾病注意情報提供支援システムによれば、疾病注意情報の提供を受ける対象者本人が疾病ごとの発症可能性の具体的な順位を直感的に理解できるので、疾病ごとの発症可能性の数値のみ、あるいは他の人を含めた疾病ごとの発症可能性の数値を一覧できるようにした統計情報を提供する場合よりも、疾病を発症している又は発症する可能性を有する者にその改善の必要性の高さを分かりやすく認識させることができる。 Thus, according to the disease attention information provision support system of this configuration, the target person who receives the provision of the disease attention information can intuitively understand the specific ranking of the onset probability for each disease. Have or will develop a disease rather than providing statistical information that allows you to list only the probability of onset or the number of possible onset for each disease, including other people Can make it easy to recognize the high necessity for improvement.
　本発明の疾病注意情報提供支援システムは、
　１又は複数の人の健康状態に関する複数の項目の値を含む健康状態情報を記憶する健康状態記憶部と、
　記憶した前記健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する発症可能性数値化部と、
　前記発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部とを備え、
　前記発症可能性数値化部は、１又は複数の人の前記健康状態情報を分析することにより疾病ごとの発症可能性と相関のある該健康状態情報の項目を特定し、対象の人の健康状態情報から前記特定された疾病ごとの発症可能性と相関のある項目の値を抽出し、該抽出された項目の値に基づいて各人の疾病ごとの発症可能性を数値化するように構成されていることを特徴とする。 The disease attention information provision support system of the present invention is
A health condition storage unit that stores health condition information including values of a plurality of items relating to the health condition of one or more people;
An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information;
An attention information generation unit that generates disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or more diseases digitized by the onset probability digitization unit,
The onset probability quantification unit identifies the item of the health status information correlated with the onset probability for each disease by analyzing the health status information of one or a plurality of people, and the health status of the target person It is configured to extract a value of an item correlated with the onset possibility for each specified disease from the information, and to quantify the onset probability for each person's disease based on the value of the extracted item. It is characterized by.
　本発明によれば、発症可能性数値化部は、１又は複数の人の前記健康状態情報を分析することにより疾病ごとの発症可能性と相関のある該健康状態情報の項目を特定する。 According to the present invention, the onset probability quantification unit identifies items of the health condition information correlated with the onset possibility for each disease by analyzing the health condition information of one or a plurality of persons.
　続いて、発症可能性数値化部は、対象の人の健康状態情報から前記特定された疾病ごとの発症可能性と相関のある項目の値を抽出する。 Subsequently, the onset probability quantification unit extracts the value of the item correlated with the onset probability for each of the identified diseases from the health status information of the target person.
　そして、発症可能性数値化部は、抽出された項目の値に基づいて各人の疾病ごとの発症可能性を数値化するように構成されている。 The onset probability quantification unit is configured to quantify the onset probability for each person's disease based on the value of the extracted item.
　各人の健康状態情報に含まれる各項目の値と、疾病ごとの発症可能性との間には、一定の相関関係があるため、健康状態情報の中から疾病ごとの発症可能性と相関のある健康状態情報の項目を特定して当該項目の値を抽出したうえで、それらの項目の値に基づいて疾病ごとの発症可能性を数値化することにより、健康状態情報のすべての項目を考慮するよりも効率的に発症可能性を数値化することができる。 Since there is a certain correlation between the value of each item included in each person's health status information and the likelihood of onset for each disease, there is a correlation between the probability of onset for each disease from the health status information. All items of health status information are taken into account by identifying the items of certain health status information and extracting the values of those items, and then quantifying the likelihood of each disease based on the values of those items The probability of onset can be quantified more efficiently than
　なお、健康状態情報に含まれるいずれの項目の値が疾病ごとの発症可能性と相関を有するかを経験則などに基づいてあらかじめ定めてしまった場合、疾病ごとの発症可能性との間に相関を有する他の重要な健康状態情報の項目を見落としてしまい、実態にそぐわない疾病ごとの発症可能性の数値を導出してしまうおそれがある。 In addition, if it is determined in advance based on empirical rules, etc. that the value of which item included in the health status information has a correlation with the probability of occurrence for each disease, it correlates with the probability of occurrence for each disease. There is a possibility that other important items of health condition information having an oversight may be overlooked, and a numerical value of the possibility of onset for each disease that does not match the actual situation may be derived.
　これに加え、健康状態情報に含まれる項目の数が多数である場合、これらの項目から疾病ごとの発症可能性との間に相関を有する項目を担当者が目視、手計算で特定することがそもそも非常に困難であることが予想される。 In addition, if the number of items included in health status information is large, the person in charge may specify items that have a correlation between these items and the likelihood of onset for each disease by visual inspection or manual calculation. In the first place, it is expected to be very difficult.
　本構成の疾病注意情報提供支援システムによれば、発症可能性数値化部は、「１又は複数の人の前記健康状態情報を分析」することにより「疾病ごとの発症可能性と相関のある該健康状態情報の項目を特定」し、さらに「対象の人の健康状態情報から前記特定された疾病ごとの発症可能性と相関のある項目の値を抽出」したうえで「抽出された項目の値に基づいて各人の疾病ごとの発症可能性を数値化」するように構成されているので、疾病ごとの発症可能性に実際に相関を持つ項目の値を考慮して疾病の発症可能性を数値化できるとともに、健康状態情報のすべての項目の値を考慮するよりも効率的に発症可能性を数値化することができる。 According to the disease attention information provision support system of this configuration, the onset probability quantification unit performs the “analysis of the health status information of one or a plurality of persons” to thereby “correlate with the onset probability for each disease. “Identify items of health condition information” and “Extract values of items correlated with the probability of occurrence of each identified disease from the health condition information of the target person” Based on the above, it is configured to `` numerize the probability of occurrence of each person's disease '', so consider the value of items that are actually correlated with the probability of occurrence of each disease, In addition to being numerical, it is possible to efficiently quantify the likelihood of onset rather than considering the values of all items of health condition information.
　本発明の疾病注意情報提供支援システムは、
　１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶する健康状態記憶部と、
　記憶した前記健康状態情報に基づいて各人の複数の疾病それぞれの発症可能性を数値化する発症可能性数値化部と、
　前記発症可能性数値化部により数値化された複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部とを備え、
　前記注意情報生成部は、前記複数の疾病のうち発症可能性の数値が所定の条件を満たす疾病を認識し、該疾病に関連付けられて記憶された合併症を認識し、該合併症に関する情報を含む前記疾病注意情報を生成するように構成されていることを特徴とする。 The disease attention information provision support system of the present invention is
A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons;
An onset probability quantification unit that quantifies the onset probability of each of a plurality of diseases of each person based on the stored health condition information;
An attention information generation unit that generates disease attention information including information indicating a numerical value of the onset probability of at least one disease among a plurality of diseases digitized by the onset probability digitization unit,
The attention information generating unit recognizes a disease whose numerical value of the probability of occurrence satisfies a predetermined condition among the plurality of diseases, recognizes a complication stored in association with the disease, and stores information on the complication It is comprised so that the said disease caution information containing may be produced | generated.
　疾病の注意情報を提供しても、専門知識を有しない人に病状の深刻さを理解させることができなければ、各人が、医療機関での受診を怠ったり、自己の健康管理を怠ったりするなどし、場合によっては健康状態が悪化してしまうおそれがある。 If people who do not have specialized knowledge cannot understand the seriousness of their medical condition even if they provide caution information about the disease, each person may fail to visit a medical institution or neglect their own health management. In some cases, the health condition may be deteriorated.
　そのような課題に対応するため、各人が発症する可能性のある疾病に関する情報に加えて、当該疾病に起因して発症する可能性のある合併症に関する情報を各人に提供することが考えられる。 In order to respond to such issues, in addition to information on diseases that each person may develop, it may be possible to provide each person with information on complications that may develop due to the disease. It is done.
　しかしながら、例えば疾病の発症可能性の有無のみに応じて、画一的に当該疾病に起因して合併症が発症する可能性がある旨を各人に提供するだけでは、各人は病状の深刻さを十分に理解できないことも懸念される。 However, simply providing each person with the possibility that complications may occur due to the disease, depending only on the presence or absence of the disease, for example, There is also a concern that this cannot be fully understood.
　本構成の疾病注意情報提供支援システムによれば、「前記複数の疾病のうち発症可能性の数値が所定の条件を満たす疾病を認識」し、さらに「該疾病に関連付けられて記憶された合併症を認識」したうえで、「該合併症に関する情報を含む前記疾病注意情報を生成」する。 According to the disease attention information provision support system of the present configuration, “recognizes a disease in which the numerical value of the probability of occurrence among the plurality of diseases satisfies a predetermined condition” and further stores “a complication associated with the disease and stored. And “generates the disease attention information including information related to the complication”.
　これにより、疾病注意情報は、各人が発症する可能性のある疾病に起因して発症する可能性のある合併症に関する情報を含んでいるので、疾病を発症している又は発症する可能性を有する者にその改善の必要性を、個別の疾病について情報提供した場合よりも強く印象付けることができる。 As a result, the disease attention information includes information on complications that may develop due to the diseases that each person may develop. The need for improvement can be impressed more strongly by those who have the information than when information about individual diseases is provided.
　また、かかる合併症に関する情報は、複数の疾病のうち発症可能性の数値が所定の条件を満たす疾病に関連付けづけられた合併症について生成されるので、各人ごとの疾病の発症可能性の数値に則した合併症に関する情報を提供することができる。 In addition, since information on such complications is generated for complications in which a numerical value of the probability of occurrence among a plurality of diseases is associated with a condition that satisfies a predetermined condition, the numerical value of the probability of developing the disease for each person Can provide information on complications.
　このように、本構成の疾病注意情報提供支援システムによれば、疾病の発症可能性の改善の必要性を個別の疾病について情報提供した場合よりも強く印象付けることができるとともに、各人ごとの疾病の発症可能性の数値に則した合併症に関する情報を提供することができる。 Thus, according to the disease attention information provision support system of this configuration, it is possible to impress the necessity of improving the possibility of developing a disease more strongly than when information is provided about individual diseases, and for each person. It is possible to provide information on complications according to the numerical value of the probability of disease occurrence.
紙媒体に印刷された疾病注意情報の一例を示す図。The figure which shows an example of the disease attention information printed on the paper medium. 疾病注意情報提供支援システムの一例の全体構成図。The whole block diagram of an example of a disease caution information provision support system. 健康状態記憶部に記憶された情報の一例を示す図。The figure which shows an example of the information memorize | stored in the health condition memory | storage part. 発症可能性相関項目記憶部に記憶された情報の一例を示す図。The figure which shows an example of the information memorize | stored in the onset possibility correlation item memory | storage part. 変化相関項目記憶部に記憶された情報の一例を示す図。The figure which shows an example of the information memorize | stored in the change correlation item memory | storage part. 予測値情報記憶部に記憶された情報の一例を示す図。The figure which shows an example of the information memorize | stored in the predicted value information storage part. 発症可能性数値情報記憶部に記憶された情報の一例を示す図。The figure which shows an example of the information memorize | stored in the onset possibility numerical value information storage part. 履歴記憶部に記憶された情報の一例を示す図。The figure which shows an example of the information memorize | stored in the log | history memory | storage part. 健康状態情報の取り出しから準備処理までの処理のフローチャート。The flowchart of the process from extraction of health condition information to a preparation process. 健康状態予測処理から疾病注意情報の出力までの処理のフローチャート。The flowchart of the process from a health condition prediction process to the output of disease caution information. 準備処理のフローチャート。The flowchart of a preparation process. 発症可能性相関項目抽出処理のフローチャート。The flowchart of onset possibility correlation item extraction processing. 発症可能性算出モデル導出処理のフローチャート。The flowchart of onset possibility calculation model derivation processing. 変化相関項目抽出処理のフローチャート。The flowchart of a change correlation item extraction process. 第２時点情報取得モデル導出処理のフローチャート。The flowchart of a 2nd time information acquisition model derivation process. 健康状態予測処理のフローチャート。The flowchart of a health condition prediction process. 予測部による第２時点における対象の人の予測される健康状態を示す情報をを取得する処理内容の一例を示す図。The figure which shows an example of the processing content which acquires the information which shows the healthy state of the object person's prediction in the 2nd time point by a prediction part. 発症可能性数値化処理のフローチャート。The flowchart of the onset possibility numerical conversion process. 発症可能性数値化部による各人の疾病ごとの発症可能性を数値化する処理内容の一例を示す図。The figure which shows an example of the processing content which digitizes the onset possibility for every disease of each person by the onset possibility digitization part. 発症可能性数値化部による各人の疾病ごとの第２時点における発症可能性を数値化する処理内容の一例を示す図。The figure which shows an example of the processing content which digitizes the onset possibility in the 2nd time for every illness of each person by the onset possibility digitization part. 注意情報生成処理のフローチャート。The flowchart of attention information generation processing. 疾病の発症可能性の数値と発症可能性の種別との変換テーブルの例を示す図。The figure which shows the example of the conversion table of the numerical value of the onset possibility of a disease, and the classification of onset possibility. 疾病の発症可能性の種別と疾病注意情報の生成頻度との変換テーブルの例を示す図。The figure which shows the example of the conversion table of the classification | category of the onset possibility of a disease, and the generation frequency of disease attention information. 各疾病の発症可能性の種別と改善アドバイスとの変換テーブブルの例を示す図。The figure which shows the example of the conversion table of the classification | category of the onset possibility of each disease, and improvement advice. 疾病と合併症との変換テーブルの例を示す図。The figure which shows the example of the conversion table of a disease and a complication.
　図１～図１４を参照しながら、本発明の疾病注意情報提供支援システムについて説明する。 The disease caution information provision support system of the present invention will be described with reference to FIGS.
　（疾病注意情報）
　疾病注意情報は、後述する注意情報生成処理（ＳＴＥＰ８０）において生成される、各人４０の疾病の発症可能性の数値と疾病の発症可能性の数値に応じて疾病の発症可能性に関する情報を含む情報である。 (Disease attention information)
The disease attention information includes information on the possibility of developing the disease according to the numerical value of the possibility of developing the disease of each person 40 and the numerical value of the possibility of developing the disease, which are generated in the attention information generation process (STEP 80) described later. Information.
　疾病注意情報は、例えば図１に示されるように、端末３０により紙媒体３２に印刷されて各人４０に対して提供される。 The disease attention information is printed on a paper medium 32 by the terminal 30 and provided to each person 40 as shown in FIG.
　あるいは疾病注意情報は、図１に示される内容を含むドキュメントファイルとして端末３０により出力され、各人４０の利用するパソコン、タブレット、スマートフォンなどの端末４１にダウンロードされることにより提供されることとしてもよい。 Alternatively, the disease warning information may be provided by being output by the terminal 30 as a document file including the contents shown in FIG. 1 and downloaded to the terminal 41 such as a personal computer, tablet, or smartphone used by each person 40. Good.
　疾病注意情報は、図１に示されるように、例えば各人４０の過去及び最も直近に取得された健康状態情報３２１、疾病ごとの発症可能性の改善方法に関する情報である改善アドバイス情報３２２、疾病の発症可能性の数値に応じて各人が発症する可能性のある疾病に起因して発症する可能性のある合併症に関する情報３２３、疾病ごとの発症可能性の数値を表すグラフ３２４、複数の人の全部又は一部により構成されるグループにおける各人の疾病ごとの発症可能性の順位に関する情報３２５、疾病ごとの前記第２時点における発症可能性の数値を含むグラフ３２６を含んでいる。 As shown in FIG. 1, the disease attention information includes, for example, the past and most recently acquired health state information 321 of each person 40, improvement advice information 322 that is information on how to improve the possibility of developing each disease, disease Information 323 relating to complications that may occur due to a disease that each person may develop according to the numerical value of the possibility of onset, a graph 324 that represents the numerical value of the probability of occurrence for each disease, It includes information 325 relating to the ranking of the probability of occurrence of each person's disease in a group composed of all or part of the person, and a graph 326 including numerical values of the probability of occurrence at the second time point for each disease.
　なお、疾病注意情報は、これらすべての情報を必ずしも含んでいる必要はなく、一部を含むこととしてもよい。また、疾病注意情報の具体的な生成方法については後述する。 It should be noted that the disease caution information does not necessarily include all the information, and may include a part. In addition, a specific method for generating disease attention information will be described later.
　（疾病注意情報提供支援システム）
　疾病注意情報提供支援システムは、１又は複数の人に対し疾病の発症可能性に関する情報を含む疾病注意情報を生成するシステムである。 (Disease attention information provision support system)
The disease attention information provision support system is a system for generating disease attention information including information on the possibility of developing a disease for one or a plurality of people.
　疾病注意情報提供支援システムは、図２に示されるように、疾病注意情報提供支援サーバ１０と、１又は複数の端末３０とを備える。 As shown in FIG. 2, the disease attention information provision support system includes a disease attention information provision support server 10 and one or more terminals 30.
　疾病注意情報提供支援サーバ１０と１又は複数の端末３０とは、ＬＡＮやインターネット等の情報通信網２０を介して相互に通信可能に構成されている。なお、図２では１つの端末３０を示している。 The disease attention information provision support server 10 and the one or more terminals 30 are configured to be able to communicate with each other via an information communication network 20 such as a LAN or the Internet. In FIG. 2, one terminal 30 is shown.
　あるいは、疾病注意情報提供支援サーバ１０が端末３０としても動作することとしてもよい。この場合は、疾病注意情報提供支援システムは、端末３０を備えない。 Alternatively, the disease attention information provision support server 10 may operate as the terminal 30. In this case, the disease attention information provision support system does not include the terminal 30.
　（疾病注意情報提供支援サーバ）
　疾病注意情報提供支援サーバ１０は、サーバ制御部１００と、サーバ記憶部１１０とを備えている。なお、疾病注意情報提供支援サーバ１０を構成するコンピュータの一部又は全部が端末３０を構成するコンピュータにより構成されていてもよい。 (Disease attention information provision support server)
The disease attention information provision support server 10 includes a server control unit 100 and a server storage unit 110. A part or all of the computer constituting the disease caution information provision support server 10 may be configured by a computer constituting the terminal 30.
　サーバ制御部１００は、ＣＰＵ（Ｃｅｎｔｒａｌ　Ｐｒｏｃｅｓｓｉｎｇ　Ｕｎｉｔ）等の演算処理装置、主記憶装置、及び入出力装置により構成されている。サーバ制御部１００は、１又は複数のプロセッサにより構成される。 The server control unit 100 includes an arithmetic processing device such as a CPU (Central Processing Unit), a main storage device, and an input / output device. The server control unit 100 is configured by one or a plurality of processors.
　サーバ制御部１００は、所定のプログラムを読み込んで実行することにより、後述の演算処理を実行する発症可能性数値化部１０１、予測部１０２、注意情報生成部１０３及び注意情報送信部１０４として機能する。 The server control unit 100 functions as an onset probability digitizing unit 101, a predicting unit 102, a caution information generating unit 103, and a caution information transmitting unit 104 that execute a calculation process described later by reading and executing a predetermined program. .
　サーバ記憶部１１０は、例えばＲＯＭ（Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ）、ＲＡＭ（Ｒａｎｄｏｍ　Ａｃｃｅｓｓ　Ｍｅｍｏｒｙ）、ＨＤＤ（Ｈａｒｄ　Ｄｉｓｋ　Ｄｒｉｖｅ）等の記憶装置により構成されている。 The server storage unit 110 includes, for example, a storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), or the like.
　サーバ記憶部１１０は、サーバ制御部１００の演算結果、又は端末３０から取り込んだ健康状態情報を記憶するように構成されている。 The server storage unit 110 is configured to store the calculation result of the server control unit 100 or the health condition information captured from the terminal 30.
　サーバ記憶部１１０は、健康状態記憶部１１１、発症可能性相関項目記憶部１１２、算出モデル記憶部１１３、変化相関項目記憶部１１４、予測モデル記憶部１１５、予測値情報記憶部１１６、発症可能性数値情報記憶部１１７、履歴記憶部１１８及び変換テーブル記憶部１１９を備える。 The server storage unit 110 includes a health state storage unit 111, an onset probability correlation item storage unit 112, a calculation model storage unit 113, a change correlation item storage unit 114, a prediction model storage unit 115, a predicted value information storage unit 116, and an onset possibility. A numerical information storage unit 117, a history storage unit 118, and a conversion table storage unit 119 are provided.
　健康状態記憶部１１１は、１又は複数の人の健康状態に関する情報を含む情報である健康状態情報を記憶している。 The health status storage unit 111 stores health status information that is information including information on the health status of one or more people.
　健康状態記憶部１１１は、例えば図３Ａに示されるように、各人４０を特定するＩＤ、健康状態情報の取得年月日、年齢、性別、身長、体重、腹囲、総コレステロール、γ－ＧＴ、尿酸値、血糖値、ＨｂＡ１ｃ、血圧、喫煙習慣、運動習慣、飲酒習慣及び治療中の疾病名で構成される、各人４０の健康状態に関する項目の情報及びそれに付随する情報のセットを１又は複数格納している。 For example, as shown in FIG. 3A, the health status storage unit 111 includes an ID for identifying each person 40, the date of acquisition of health status information, age, sex, height, weight, waist circumference, total cholesterol, γ-GT, One or more sets of information on items related to the health status of each person 40 and information associated therewith composed of uric acid levels, blood glucose levels, HbA1c, blood pressure, smoking habits, exercise habits, drinking habits, and disease names being treated Storing.
　なお健康状態記憶部１１１は、同一のＩＤについて異なる取得年月日における当該情報のセットを複数格納しうる。 The health condition storage unit 111 can store a plurality of sets of the information for different acquisition dates for the same ID.
　なお健康状態記憶部１１１は、取得年月日がある一定の間隔（たとえば１年間）があいた情報セットを格納しうる。この１年間とは、厳密な意味での１年間のみならず、たとえば１０か月～１４か月のようにある程度幅を持った期間であってもよいし、取得年度または取得年が異なる情報セットであってもよい。 Note that the health state storage unit 111 can store an information set having a certain interval (for example, one year) with a date of acquisition. This one year is not limited to one year in a strict sense, but may be a period with a certain range, for example, 10 months to 14 months, or an information set having different acquisition years or acquisition years. It may be.
　以下、同一のＩＤについての当該情報のセットのうち、最も新しい取得年月日に取得された情報セットを「最新の健康状態情報」といい、２番目に新しい取得年月日に取得された情報セットであって当該取得年月日が最新の健康状態情報の取得年月日の所定期間前（たとえば１０カ月～１４カ月前）の範囲に含まれる情報セットを「１年前の健康状態情報」といい、３番目に新しい取得年月日に取得された情報セットであって当該取得年月日が１年間の健康状態情報の取得年月日の所定期間前（たとえば１０カ月～１４カ月前）の範囲に含まれる情報セットを「２年前の健康状態情報」という。 Hereinafter, the information set acquired on the most recent acquisition date among the set of the information for the same ID is referred to as “latest health status information”, and the information acquired on the second new acquisition date. An information set that is included in a set period before the acquisition date of the latest health condition information (for example, 10 months to 14 months before) is set as “health condition information for one year ago” The information set acquired on the third most recent acquisition date, and the acquisition date is one year before the specified period (for example, 10 to 14 months before). The information set included in the range is referred to as “health state information two years ago”.
　以下においては、上述した最新、１年前及び２年前の健康状態情報を使う場合の例を説明するが、これらに限られず、取得年月日がある程度の間隔があいた情報セットであればよい。取得時期が異なる情報セットの数は、３つに限られず、２つであってもよいし、４つ以上であってもよい。また、各人４０のそれぞれについて、その情報セットの取得時期は同一であってもよいし、異なっていてもよい。また、ある一人の人について取得時期が異なる情報セットが、最新、１年前及び２年前の健康状態情報と、１年前、２年前及び３年前の健康状態情報とのように、複数用いられてもよい。 In the following, an example of using the above-mentioned latest, 1-year-old and 2-year-old health condition information will be described. However, the present invention is not limited thereto, and any information set may be used as long as the acquisition date has a certain interval. . The number of information sets with different acquisition times is not limited to three, and may be two or four or more. In addition, for each person 40, the acquisition time of the information set may be the same or different. In addition, information sets with different acquisition times for a single person are the latest, 1 year ago, 2 years ago health status information, 1 year ago, 2 years ago and 3 years ago health status information, A plurality may be used.
　健康状態記憶部１１１は、健康状態情報を、病院又は健康保険組合等の端末３０から、ＣＤ－ＲＯＭ、ＤＶＤ－ＲＯＭ、ＵＳＢメモリなどの外部記憶媒体３１を介して、あるいは情報通信網２０を介して取り込む。 The health status storage unit 111 receives health status information from a terminal 30 such as a hospital or a health insurance association via an external storage medium 31 such as a CD-ROM, DVD-ROM, or USB memory, or via the information communication network 20. Capture.
　発症可能性相関項目記憶部１１２は、図３Ｂに示されるように、疾病名及び当該疾病の発症可能性と相関のある健康状態に関する項目のセットを１又は複数格納している。 As shown in FIG. 3B, the onset probability correlation item storage unit 112 stores one or a plurality of sets of items related to the health status correlated with the disease name and the onset possibility of the disease.
　算出モデル記憶部１１３は、後述する発症可能性算出モデル準備処理（ＳＴＥＰ４０２）において発症可能性数値化部１０１が導出した、疾病ごとの発症可能性を算出するモデルを格納している。 The calculation model storage unit 113 stores a model for calculating the onset probability for each disease derived by the onset probability digitizing unit 101 in the onset possibility calculation model preparation process (STEP 402) described later.
　変化相関項目記憶部１１４は、図３Ｃに示されるように、健康状態情報の一部またはすべての項目名及び当該項目それぞれの値の変化の大きさに相関のある健康状態情報の項目のセットを１又は複数格納している。 As shown in FIG. 3C, the change correlation item storage unit 114 stores a set of items of health state information correlated with a part or all of the item names of the health state information and the magnitude of change in the value of each item. One or more are stored.
　予測モデル記憶部１１５は、後述する第２時点情報取得モデル準備処理（ＳＴＥＰ４０４）において予測部１０２が導出した、第１時点における健康状態情報を含む情報と、当該第１時点における健康状態情報と当該第１時点より前の時点の健康状態情報との間における変化の大きさを含む情報とを入力として当該第１時点より後の第２時点における健康状態を特定できる情報を出力とするモデルを１又は複数格納している。 The prediction model storage unit 115 includes information including the health state information at the first time point, the health state information at the first time point, and the information derived by the prediction unit 102 in the second time point information acquisition model preparation process (STEP 404) described later. A model that outputs information that can specify a health state at a second time point after the first time point by inputting information including the magnitude of change between the health state information at a time point before the first time point and 1 Or a plurality are stored.
　なお、本実施例において、第１時点は、各人４０の最新の健康状態情報の取得年月日であり、第２時点は、例えば第１時点の１年後であり、第１時点より前の時点は、各人４０の１年前の健康状態情報の取得年月日である。 In the present embodiment, the first time point is the date of acquisition of the latest health information of each person 40, and the second time point is, for example, one year after the first time point and before the first time point. Is the date of acquisition of the health status information of each person 40 one year ago.
　予測値情報記憶部１１６は、図３Ｄに示されるように、後述する健康状態予測処理（ＳＴＥＰ５０）において予測部１０２が算出した、第２時点の年月日における各人４０の健康状態を示す数値及びそれに付随する情報のセットを１又は複数格納している。 As shown in FIG. 3D, the predicted value information storage unit 116 is a numerical value indicating the health status of each person 40 at the second time point calculated by the prediction unit 102 in the health status prediction process (STEP 50) described later. And one or more sets of information associated therewith.
　なお、予測値情報記憶部１１６は、同一のＩＤについて第２時点の１年後である第３時点の年月日における健康状態を示す数値、さらに１年後である第４時点の年月日における健康状態を示す数値など、第２時点以降の将来の年月日における健康状態情報を示す数値を複数格納しうる。 Note that the predicted value information storage unit 116 stores a numerical value indicating the health status of the third time point that is one year after the second time point for the same ID, and the date of the fourth time point that is one year later. It is possible to store a plurality of numerical values indicating health state information in the future date after the second time point, such as a numerical value indicating the health state in.
　発症可能性数値情報記憶部１１７は、図３Ｅに示されるように、発症可能性数値化処理（ＳＴＥＰ６０１）において発症可能性数値化部１０１が算出又は設定した糖尿病リスク、脂質異常症リスクなどの、各人４０の疾病ごとの発症可能性の数値、糖尿病リスク種別、脂質異常症リスク種別などの、各人４０の疾病ごとの発症可能性の数値に応じた種別を分類した情報及びこれらに付随する情報のセットである。 As shown in FIG. 3E, the onset possibility numerical value information storage unit 117 is calculated or set by the onset possibility number conversion unit 101 in the onset possibility numerical value calculation process (STEP 601), such as diabetes risk, dyslipidemia risk, Information that classifies the type according to the numerical value of the onset probability for each person's 40 disease, such as the numerical value of the onset possibility for each person's 40 disease, the diabetes risk type, the dyslipidemia risk type, and the like. A set of information.
　発症可能性数値情報記憶部１１７は、当該情報のセットを１又は複数格納している。 The onset probability numerical information storage unit 117 stores one or more sets of the information.
　なお発症可能性数値情報記憶部１１７は、同一のＩＤについて異なる取得年月日における当該情報のセットを複数格納しうる。 The onset probability numerical information storage unit 117 can store a plurality of sets of the information for different acquisition dates for the same ID.
　履歴記憶部１１８は、図３Ｆに示されるように、各人４０ごとの最も新しい疾病注意情報が生成された年月日である注意情報最終生成年月日及びそれに付随する情報のセットを１又は複数格納している。 As shown in FIG. 3F, the history storage unit 118 sets one or more sets of the caution information last generation date and the information associated therewith as the date of generation of the latest disease caution information for each person 40. Multiple items are stored.
　変換テーブル記憶部１１９は、図１４Ａ～図１４Ｄに示されるように、疾病注意情報提供支援サーバ１０が各種の処理を実行する際に参照する変換テーブルを格納している。 As shown in FIGS. 14A to 14D, the conversion table storage unit 119 stores conversion tables to be referred to when the disease attention information provision support server 10 executes various processes.
　（端末）
　端末３０は、デスクトップコンピュータ、タブレット型端末、スマートフォンなどにより構成される。 (Terminal)
The terminal 30 is configured by a desktop computer, a tablet terminal, a smartphone, or the like.
　端末３０は、例えばＲＯＭ（Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ）、ＲＡＭ（Ｒａｎｄｏｍ　Ａｃｃｅｓｓ　Ｍｅｍｏｒｙ）、ＨＤＤ（Ｈａｒｄ　Ｄｉｓｋ　Ｄｒｉｖｅ）等の記憶装置を備えており、１又は複数の人の健康状態情報及び疾病注意情報提供支援サーバ１０から受信した疾病注意情報が記憶されている。 The terminal 30 includes a storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), etc., and a server for providing health status information and disease attention information for one or more people. The disease attention information received from 10 is stored.
　端末３０は、例えば図１に示されるように、紙媒体３２に印刷して各人４０に対して、疾病注意情報を提供する。 For example, as shown in FIG. 1, the terminal 30 prints on a paper medium 32 and provides disease warning information to each person 40.
　あるいは、端末３０は、図１に示される内容を含むドキュメントファイルを出力し、当該ドキュメントファイルを各人４０の利用するパソコン、タブレット、スマートフォンなどの端末４１にダウンロードさせることにより疾病注意情報を提供することとしてもよい。 Alternatively, the terminal 30 outputs a document file including the contents shown in FIG. 1 and provides the disease warning information by causing the terminal 41 such as a personal computer, a tablet, or a smartphone used by each person 40 to download the document file. It is good as well.
　あるいは、疾病注意情報提供支援サーバ１０が、疾病注意情報の紙媒体３２への印刷又はドキュメントファイルの出力、及び各人への提供を行うこととしてもよい。 Alternatively, the disease caution information provision support server 10 may print the disease caution information on a paper medium 32 or output a document file and provide it to each person.
　（処理の概要）
　図４を参照して、疾病注意情報提供支援システムの処理の概要について説明する。 (Outline of processing)
With reference to FIG. 4, the outline | summary of the process of a disease caution information provision assistance system is demonstrated.
　疾病注意情報提供支援システムは、図４Ａに示される、端末３０による健康状態情報の取り出し（ＳＴＥＰ１０）から疾病注意情報提供支援サーバ１０による準備処理（ＳＴＥＰ４０）までの処理と、図４Ｂに示される、疾病注意情報提供支援サーバ１０による健康状態予測処理（ＳＴＥＰ５０）から端末３０による疾病注意情報の出力（ＳＴＥＰ１２０）までの処理とを実行する。 The disease caution information provision support system is illustrated in FIG. 4A, from the health condition information extraction (STEP 10) by the terminal 30 to the preparation process (STEP 40) by the disease caution information provision support server 10, as shown in FIG. 4B. The process from the health state prediction process (STEP 50) by the disease attention information provision support server 10 to the output of the disease attention information (STEP 120) by the terminal 30 is executed.
　なお、疾病注意情報提供支援サーバ１０又は端末３０が健康状態情報又は健康状態情報の項目について処理を行う場合において、疾病注意情報提供支援サーバ１０又は端末３０が処理対象とする健康状態情報又は健康状態情報の項目は、その一部又は全部である。 In addition, in the case where the disease attention information provision support server 10 or the terminal 30 performs processing on the health condition information or the item of the health condition information, the health condition information or the health condition to be processed by the disease attention information provision support server 10 or the terminal 30 The item of information is a part or all of the item.
　図４Ａに示される処理は、各人４０の健康状態情報を蓄積するとともに図４Ｂに示される一連の処理の実行のための準備をする一連の処理であり、図４Ｂに示される処理は、疾病注意情報の生成及び送信のための一連の処理である。 The process shown in FIG. 4A is a series of processes for accumulating the health condition information of each person 40 and preparing for the execution of the series of processes shown in FIG. 4B. The process shown in FIG. It is a series of processes for generating and transmitting attention information.
　端末３０及び疾病注意情報提供支援サーバ１０は、図４Ａに示される一連の処理と図４Ｂに示される一連の処理とを、そのすべてを一連の処理として一度に行ってもよいし、異なる頻度で行ってもよい。 The terminal 30 and the disease attention information provision support server 10 may perform the series of processes shown in FIG. 4A and the series of processes shown in FIG. 4B all at once as a series of processes, or at different frequencies. You may go.
　まず、図４Ａに示される、端末３０による健康状態情報の取り出し（ＳＴＥＰ１０）から疾病注意情報提供支援サーバ１０による準備処理（ＳＴＥＰ４０）までの処理について説明する。 First, the process from the health condition information extraction by the terminal 30 (STEP 10) to the preparation process (STEP 40) by the disease attention information provision support server 10 shown in FIG. 4A will be described.
　まず端末３０は、１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶装置から取り出す（ＳＴＥＰ１０）。当該取り出しは、ＣＤ－ＲＯＭ、ＤＶＤ－ＲＯＭ、ＵＳＢメモリなどの外部記憶媒体３１に健康状態情報を格納することにより行われてもよいし、あるいは、情報通信網２０を介して送信先を特定して健康状態情報を送信することにより行われてもよい。 First, the terminal 30 takes out health status information including information on the health status of one or more persons from the storage device (STEP 10). The retrieval may be performed by storing health state information in an external storage medium 31 such as a CD-ROM, DVD-ROM, or USB memory, or by specifying a transmission destination via the information communication network 20. It may be done by sending health status information.
　次に、疾病注意情報提供支援サーバ１０は、端末３０が取り出した健康状態情報を取り込み（ＳＴＥＰ２０）、健康状態記憶部１１１に記憶する（ＳＴＥＰ３０）。 Next, the disease caution information provision support server 10 takes in the health condition information taken out by the terminal 30 (STEP 20) and stores it in the health condition storage unit 111 (STEP 30).
　その後、疾病注意情報提供支援サーバ１０は準備処理（ＳＴＥＰ４０）を実行し、健康状態情報の予測や疾病の発症可能性の数値化に必要な準備を行う。準備処理（ＳＴＥＰ４０）の詳細については後述する。 Thereafter, the disease attention information provision support server 10 executes a preparation process (STEP 40), and makes preparations necessary for predicting the health condition information and quantifying the possibility of developing the disease. Details of the preparation process (STEP 40) will be described later.
　端末３０及び疾病注意情報提供支援サーバ１０は、図４Ａに示される一連の処理を、そのすべてを一連の処理として一度に行ってもよいし、あるいは例えば、ＳＴＥＰ１０～３０のみを毎回行い、データが一定量蓄積されるごとや一定期間ごとにＳＴＥＰ４０を実行することとしてもよい。 The terminal 30 and the disease attention information provision support server 10 may perform the series of processes shown in FIG. 4A all at once as a series of processes. Alternatively, for example, only the STEPs 10 to 30 are performed each time, and the data is stored. STEP 40 may be executed every time a certain amount is accumulated or at regular intervals.
　次に、図４Ｂに示される、疾病注意情報提供支援サーバ１０による健康状態予測処理（ＳＴＥＰ５０）から端末３０による疾病注意情報の出力（ＳＴＥＰ１２０）までの処理について説明する。 Next, processing from the health state prediction process (STEP 50) by the disease attention information provision support server 10 to the output of disease attention information by the terminal 30 (STEP 120) shown in FIG. 4B will be described.
　まず、予測部１０２が健康状態予測処理（ＳＴＥＰ５０）を行う。 First, the prediction unit 102 performs a health state prediction process (STEP 50).
　次に、発症可能性数値化部１０１が発症可能性数値化処理（ＳＴＥＰ６０）を実行して各人４０の疾病の発症可能性を数値化し、発症可能性数値情報を発症可能性数値情報記憶部１１７に記憶する（ＳＴＥＰ７０）。 Next, the onset probability numerical unit 101 executes the onset probability digitization process (STEP 60) to digitize the onset probability of the disease of each person 40, and the onset probability numerical information is stored as the onset probability numerical information storage unit. It memorize | stores in 117 (STEP70).
　続いて、注意情報生成部１０３が注意情報生成処理（ＳＴＥＰ８０）を実行して各人４０の少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成し、注意情報送信部１０４が、発症可能性数値情報を含む疾病注意情報を端末３０に送信する（ＳＴＥＰ９０）。 Subsequently, the attention information generation unit 103 executes attention information generation processing (STEP 80) to generate disease attention information including information indicating the numerical value of at least one disease onset of each person 40, and the attention information transmission unit 104 transmits the disease attention information including onset probability numerical value information to the terminal 30 (STEP 90).
　なお、疾病の発症可能性の数値を示す情報は、疾病の発症可能性の数値を明確に示す情報のみならず、疾病の発症可能性の数値を大まかに（概略的に）示す情報であってもよい。例えば、疾病の発症可能性の数値を示す情報は、疾病の発症可能性の数値自体のみならず、疾病の発症可能性の数値を表すグラフ、疾病の発症可能性の数値を表す色等の情報であってもよい。 Note that the information indicating the numerical value of the probability of developing the disease is not only information clearly indicating the numerical value of the probability of developing the disease, but also information that roughly (schematically) indicates the numerical value of the probability of developing the disease. Also good. For example, the information indicating the numerical value of the probability of developing the disease includes not only the numerical value of the probability of developing the disease itself but also information such as a graph indicating the numerical value of the probability of developing the disease and the color indicating the numerical value of the probability of developing the disease. It may be.
　健康状態予測処理（ＳＴＥＰ５０）、発症可能性数値化処理（ＳＴＥＰ６０）及び注意情報生成処理（ＳＴＥＰ８０）の詳細については後述する。 Details of the health state prediction process (STEP 50), the onset probability digitization process (STEP 60), and the attention information generation process (STEP 80) will be described later.
　端末３０は、疾病注意情報提供支援サーバ１０から送信された発症可能性数値情報を含む疾患注意情報を受信して（ＳＴＥＰ１００）、記憶し（ＳＴＥＰ１１０）、出力をして（ＳＴＥＰ１２０）処理は終了する。 The terminal 30 receives (step 100), stores (STEP 110), outputs (STEP 120), and outputs (STEP 120) processing of the disease attention information including the numerical value information about the possibility of onset transmitted from the disease attention information provision support server 10. .
　端末３０及び疾病注意情報提供支援サーバ１０は、図４Ｂに示される一連の処理を、そのすべてを一連の処理として一度に行ってもよいし、あるいは例えば、ＳＴＥＰ５０～１１０までを毎回行い、データが一定量蓄積されるごとや一定期間ごとにＳＴＥＰ１２０を実行することとしてもよい。 The terminal 30 and the disease attention information provision support server 10 may perform the series of processes shown in FIG. 4B all at once as a series of processes, or, for example, perform steps 50 to 110 every time, STEP 120 may be executed every time a certain amount is accumulated or at regular intervals.
　（準備処理）
　準備処理（ＳＴＥＰ４０）は、変化相関項目抽出処理（ＳＴＥＰ４０１）、第２時点情報取得モデル準備処理（ＳＴＥＰ４０２）、発症可能性相関項目抽出処理（ＳＴＥＰ４０３）及び発症可能性算出モデル準備処理（ＳＴＥＰ４０４）により構成される。 (Preparation process)
The preparation process (STEP 40) includes a change correlation item extraction process (STEP 401), a second time point information acquisition model preparation process (STEP 402), an onset possibility correlation item extraction process (STEP 403), and an onset possibility calculation model preparation process (STEP 404). Composed.
　以下、図６及び図７を用いて、それぞれの処理の詳細について説明する。 Hereinafter, the details of each process will be described with reference to FIGS. 6 and 7.
　なお、疾病注意情報提供支援サーバ１０は、これらの処理を、そのすべてを一連の処理として一度に行ってもよいし、それぞれを異なる頻度で行ってもよい。 The disease caution information provision support server 10 may perform all of these processes at once as a series of processes, or may perform each of them at a different frequency.
　（変化相関項目抽出処理）
　変化相関項目抽出処理（ＳＴＥＰ４０１）は、健康状態情報の各項目の値の変化の大きさと相関のある健康状態情報の項目を健康状態情報の分析により抽出する処理である。処理内容は図６Ａに示される通りであり、健康状態情報に含まれる項目ごと（ループＬ１）に行われる。 (Change correlation item extraction processing)
The change correlation item extraction process (STEP 401) is a process of extracting items of health state information correlated with the magnitude of change in the value of each item of health state information by analyzing the health state information. The processing content is as shown in FIG. 6A and is performed for each item (loop L1) included in the health condition information.
　以下、健康状態情報の各項目の１年間の値の変化の大きさと相関のある健康状態情報の項目を抽出する場合について説明する。 Hereinafter, the case of extracting items of health status information correlated with the magnitude of the change in the value of each year of the health status information items will be described.
　まず、予測部１０２は、健康状態記憶部１１１に記憶されている複数の人の最新の健康状態情報の各項目の値と、１年前の健康状態情報の各項目の値と、２年前の健康状態情報の各項目の値とを取得する（ＳＴＥＰ４０１１）。 First, the prediction unit 102 calculates the value of each item of the latest health state information of a plurality of persons stored in the health state storage unit 111, the value of each item of health state information one year ago, and two years ago. The value of each item of the health status information is acquired (STEP 4011).
　次に、予測部１０２は、取得した最新の健康状態情報の各項目の値と１年前の健康状態情報の各項目の値との差分を当該複数の人について計算する（ＳＴＥＰ４０１２）。 Next, the prediction unit 102 calculates the difference between the value of each item of the acquired latest health condition information and the value of each item of the health condition information one year ago for the plurality of persons (STEP 4012).
　つまり例えば、ある人の最新の健康状態情報の値が、体重が８９、腹囲が８０、総コレステロールが１５０、血糖値が１２５であり、１年前の値が、体重が８６、腹囲が７８、総コレステロールが１５５、血糖値が１２４である場合、最新の健康状態情報の値と１年前の健康状態情報の値との差分は、体重が＋３、腹囲が＋２、総コレステロールが－５、血糖値が＋１となる。 That is, for example, the value of the latest health condition information of a person is weight 89, abdominal circumference 80, total cholesterol 150, blood sugar level 125, the value of one year ago is weight 86, abdominal circumference 78, If the total cholesterol is 155 and the blood glucose level is 124, the difference between the value of the latest health information and the value of the health information one year ago is +3 for body weight, +2 for waist circumference, -5 for total cholesterol, blood glucose The value is +1.
　続いて予測部１０２は、取得した１年前の健康状態情報の各項目の値と２年前の健康状態情報の各項目の値との差分を当該複数の人について計算する（ＳＴＥＰ４０１３）。 Subsequently, the prediction unit 102 calculates the difference between the value of each item of the acquired health condition information one year ago and the value of each item of the health condition information two years ago for the plurality of persons (STEP 4013).
　予測部１０２は、ＳＴＥＰ４０１２と同様の計算により、取得した１年前の健康状態情報の値と２年前の健康状態情報の値との差分を計算する。 The prediction unit 102 calculates the difference between the acquired value of the health condition information one year ago and the value of the health condition information two years ago by the same calculation as STEP 4012.
　次に、予測部１０２は、値の変化の大きさと相関のある健康状態情報の項目の抽出対象である健康状態情報の項目について、当該項目の値の変化の大きさに対する健康状態情報の各項目の相関の度合いを算出する（ＳＴＥＰ４０１４）。予測部１０２は、相関の度合いの算出を、例えば線形回帰分析により行う。 Next, the prediction unit 102 selects each item of the health state information with respect to the magnitude of the change in the value of the item for the health state information that is the extraction target of the item of the health state information correlated with the magnitude of the change in the value. Is calculated (STEP 4014). The prediction unit 102 calculates the degree of correlation, for example, by linear regression analysis.
　すなわち例えば、腹囲の値の変化の大きさに対する各項目の相関の度合いを算出する場合においては、予測部１０２は、対象の人の腹囲の変化の大きさである１年前の健康状態情報と最新の健康状態間との間における腹囲の差分を目的変数とし、当該人の２年前の健康状態情報と１年前の健康状態間との間における腹囲の差分、当該人の１年前の腹囲の値及び当該人の１年前の健康状態情報に含まれる各項目の値を説明変数としたモデルを用いて、各項目の係数を線形回帰分析により求める。 That is, for example, in the case of calculating the degree of correlation of each item with the magnitude of the change in the abdominal circumference value, the prediction unit 102 calculates the health status information of one year ago that is the magnitude of the change in the circumference of the subject person. The difference in abdominal circumference between the latest health status is the objective variable, and the difference in the abdominal circumference between the health status information of the person two years ago and the health status of the previous year, The coefficient of each item is obtained by linear regression analysis using a model in which the value of each item contained in the abdominal circumference value and the value of each item included in the health status information of the person one year ago is an explanatory variable.
　上記のような分析によって得られる健康状態情報各項目の係数の絶対値の大きいものが、腹囲の値の変化の大きさに対する相関の度合いが高い健康状態情報の項目（例えば喫煙習慣、運動習慣、飲酒習慣、年齢及び性別）である。 Items with a large absolute value of the coefficient of each health condition information item obtained by the above analysis are items of health condition information having a high degree of correlation with the magnitude of change in the abdominal circumference value (for example, smoking habits, exercise habits, Drinking habits, age and gender).
　続いて予測部１０２は、健康状態情報の項目から、相関の度合いの高い項目を一定数抽出する（ＳＴＥＰ４０１５）。予測部１０２は、例えば、健康状態情報の項目から、相関の度合いの高い順に１０項目を選び出すなどの方法により相関の度合いの高い項目を抽出する。 Subsequently, the prediction unit 102 extracts a certain number of items having a high degree of correlation from the items of health state information (STEP 4015). For example, the prediction unit 102 extracts items having a high degree of correlation by a method such as selecting 10 items from the items of the health condition information in the order of the degree of the correlation.
　そして予測部１０２は、例えば図３Ｃに示されるような形式にて、対象の健康状態情報の項目名を予測対象として、抽出された項目名を当該健康状態情報の項目の値の変化と相関のある項目としてそれぞれ変化相関項目記憶部１１４に記憶する（ＳＴＥＰ４０１６）。 Then, the prediction unit 102 correlates the extracted item name with the change in the value of the item of the health condition information, with the item name of the target health condition information as the prediction target, for example, in a format as shown in FIG. 3C. Each item is stored in the change correlation item storage unit 114 as a certain item (STEP 4016).
　予測部１０２は、健康状態情報のすべての項目についてのこれらの処理（ＳＴＥＰ４０１１～４０１６）を終えたときにはループＬ１を抜け、変化相関項目抽出処理を終了する。 The prediction unit 102 exits the loop L1 after completing these processes (STEPs 4011 to 4016) for all items of the health condition information, and ends the change correlation item extraction process.
　（第２時点情報取得モデル準備処理）
　第２時点情報取得モデル準備処理（ＳＴＥＰ４０２）は、任意の時点における任意の人の健康状態情報を含む情報を入力として当該時点より後の時点における当該人の健康状態を示す情報を出力とするモデルである第２時点情報取得モデルに用いる定数及び健康状態情報の各項目の係数を導出し、当該モデルを生成又は更新する処理である。 (Second time point information acquisition model preparation process)
The second time point information acquisition model preparation process (STEP 402) is a model in which information including the health state information of an arbitrary person at an arbitrary time point is input and information indicating the health state of the person at a later time point is output. This is a process of deriving the constants used for the second time point information acquisition model and the coefficient of each item of the health condition information, and generating or updating the model.
　処理内容は図６Ｂに示される通りであり、健康状態情報の項目ごと（ループＬ２）に行われる。 The processing content is as shown in FIG. 6B, and is performed for each item of health status information (loop L2).
　まず、予測部１０２は、健康状態記憶部１１１に記憶されている複数の人の最新の健康状態情報の各項目の値と、１年前の健康状態情報の各項目の値と、２年前の健康状態情報の各項目の値とを取得する（ＳＴＥＰ４０２１）。 First, the prediction unit 102 calculates the value of each item of the latest health state information of a plurality of persons stored in the health state storage unit 111, the value of each item of health state information one year ago, and two years ago. The value of each item of the health condition information is acquired (STEP 4021).
　次に、予測部１０２は、取得した最新の健康状態情報の各項目の値と１年前の健康状態情報の各項目の値との差分を当該複数の人について計算する（ＳＴＥＰ４０２２）。 Next, the prediction unit 102 calculates the difference between the value of each item of the acquired latest health condition information and the value of each item of the health condition information one year ago for the plurality of persons (STEP 4022).
　続いて予測部１０２は、取得した１年前の健康状態情報の各項目の値と２年前の健康状態情報の各項目の値との差分を当該複数の人について計算する（ＳＴＥＰ４０２３）。 Subsequently, the prediction unit 102 calculates a difference between the value of each item of the acquired health condition information one year ago and the value of each item of the health condition information two years ago for the plurality of persons (STEP 4023).
　次に、予測部１０２は、変化相関項目記憶部１１４に記憶された情報を参照し、対象の項目の値の変化の大きさと相関のある健康状態情報の項目名を取得する（ＳＴＥＰ４０２４）。 Next, the prediction unit 102 refers to the information stored in the change correlation item storage unit 114, and acquires the item name of the health state information correlated with the magnitude of the change in the value of the target item (STEP 4024).
　すなわち例えば、対象の項目が腹囲である場合、予測部１０２は、健康状態情報の項目の値の変化と相関のある項目として変化相関項目記憶部１１４に記憶された健康状態情報の項目名が喫煙習慣、運動習慣、飲酒習慣、年齢及び性別であることを認識する。 That is, for example, when the target item is abdominal circumference, the prediction unit 102 determines that the item name of the health condition information stored in the change correlation item storage unit 114 as the item correlated with the change in the value of the health condition information item is smoking. Recognize habits, exercise habits, drinking habits, age and gender.
　次に、予測部１０２は、対象の健康状態情報の項目の第２時点情報取得モデルに用いる定数及び健康状態情報の各項目の係数を導出する（ＳＴＥＰ４０２５）。このモデルは、変化相関項目抽出処理（ＳＴＥＰ４０１）におけるモデルと同種であり、例えば線形回帰モデルである。 Next, the prediction unit 102 derives constants used for the second time point information acquisition model of the target health condition information item and coefficients of each item of the health condition information (STEP 4025). This model is the same type as the model in the change correlation item extraction process (STEP 401), and is, for example, a linear regression model.
　すなわち例えば、腹囲の第２時点情報取得モデルを生成又は更新する場合においては、予測部１０２は、対象の人の腹囲の変化の大きさである１年前の健康状態情報と最新の健康状態間との間における腹囲の差分を目的変数とし、当該人の２年前の健康状態情報と１年前の健康状態間との間における腹囲の差分、当該人の１年前の腹囲の値及び当該人の１年前の健康状態情報に含まれる項目のうち腹囲の変化の大きさと相関のある項目の値を説明変数としたモデルの、定数及び各項目の係数を線形回帰分析により求める。 That is, for example, in the case of generating or updating the second time point information acquisition model of the abdominal circumference, the prediction unit 102 determines whether the health status information of one year ago that is the magnitude of the change in the abdominal circumference of the target person and the latest health status. The difference in abdominal circumference between the health status information of the person 2 years ago and the health status of the previous year, the value of the abdominal circumference of the person 1 year ago, and the A constant and a coefficient of each item of a model in which the value of the item correlated with the magnitude of the change in the abdominal circumference among items included in the health condition information of one year ago is obtained by linear regression analysis.
　予測部１０２は、このような処理により求めた定数、各項目の係数を線形回帰分析の式に当てはめることにより、対象の健康状態情報の項目の第２時点情報取得モデルを生成する（ＳＴＥＰ４０２６）。 The prediction unit 102 generates the second time point information acquisition model of the item of the health condition information of the target by applying the constant obtained by such processing and the coefficient of each item to the equation of the linear regression analysis (STEP 4026).
　予測部１０２は、生成した健康状態の各項目の第２時点情報取得モデルを予測モデル記憶部１１５に記憶する（ＳＴＥＰ４０２７）。 The prediction unit 102 stores the generated second time point information acquisition model of each item of the health condition in the prediction model storage unit 115 (STEP 4027).
　なお、すでに健康状態の当該項目の第２時点情報取得モデルが予測モデル記憶部１１５に記憶されている場合は、予測部１０２は、生成した第２時点情報取得モデルで上書きをして記憶することで、健康状態の当該項目の第２時点情報取得モデルを更新する。 When the second time point information acquisition model of the item in the health state is already stored in the prediction model storage unit 115, the prediction unit 102 overwrites and stores the generated second time point information acquisition model. Then, the second time point information acquisition model of the item of the health condition is updated.
　予測部１０２は、健康状態情報のすべての項目についてのこれらの処理（ＳＴＥＰ４０２１～４０２７）を終えたときにはループＬ２を抜け、第２時点情報取得モデル準備処理を終了する。 The prediction unit 102 exits the loop L2 after completing these processes (STEPs 4021 to 4027) for all items of the health condition information, and ends the second time point information acquisition model preparation process.
　（発症可能性相関項目抽出処理）
　発症可能性相関項目抽出処理（ＳＴＥＰ４０３）は、各疾病の発症可能性と相関のある項目を健康状態情報の分析により抽出する処理である。処理内容は、図７Ａに示される通りであり、発症可能性相関項目記憶部１１２に記憶されている疾病ごと（ループＬ３）に行われる。 (Potential correlation item extraction processing)
The onset probability correlation item extraction process (STEP 403) is a process of extracting items having a correlation with the onset possibility of each disease by analyzing health condition information. The processing content is as shown in FIG. 7A and is performed for each disease (loop L3) stored in the onset possibility correlation item storage unit 112.
　まず、発症可能性数値化部１０１は、健康状態記憶部１１１に記憶されている複数の人の健康状態情報から、対象の疾病の発症有無が分かる人の最新の健康状態情報を抽出する（ＳＴＥＰ４０３１）。 First, the onset probability quantification unit 101 extracts the latest health state information of a person who can know the onset of the target disease from the health state information of a plurality of people stored in the health state storage unit 111 (STEP 4031). ).
　発症可能性数値化部１０１は、疾病の発症有無が分かるか否かを、例えば健康状態情報に含まれる治療中の疾病名を参照して判定する。 The onset probability quantification unit 101 determines whether or not the onset of the disease is known with reference to, for example, the name of the disease under treatment included in the health condition information.
　すなわち例えば、治療中の疾病名に何らかの疾病名がある場合及び治療中の疾病名がないことが明確に記載されている場合は疾病の発症有無が分かるものと判定し、治療中の疾病名に値が存在しない場合は、疾病の発症有無が分からないと判定する。 That is, for example, if there is any disease name in the name of the disease being treated or if it is clearly stated that there is no disease name being treated, it is determined that the presence or absence of the disease is known, and the disease name being treated If no value is present, it is determined that the presence or absence of the disease is unknown.
　次に、発症可能性数値化部１０１は、対象の疾病の発症に対する健康状態情報の各項目の相関の度合いを算出する（ＳＴＥＰ４０３２）。発症可能性数値化部１０１は、例えばロジスティック回帰分析により相関の度合いを算出する。 Next, the onset probability digitization unit 101 calculates the degree of correlation of each item of the health status information with respect to the onset of the target disease (STEP 4032). The onset probability digitization unit 101 calculates the degree of correlation by, for example, logistic regression analysis.
　すなわち例えば、糖尿病の発症に対する各項目の相関の度合いを算出する場合においては、発症可能性数値化部１０１は、対象の人の糖尿病の発症確率を目的変数とし、当該人の健康状態情報に含まれる各項目の値を説明変数としたモデルの、各項目の偏回帰係数をロジスティック回帰分析により求める。 That is, for example, in the case of calculating the degree of correlation of each item with respect to the onset of diabetes, the onset probability quantification unit 101 uses the onset probability of diabetes of the target person as an objective variable and is included in the health status information of the person The partial regression coefficient of each item of the model using the value of each item as an explanatory variable is obtained by logistic regression analysis.
　上記のような分析によって得られる健康状態情報の各項目の偏回帰係数の絶対値の大きいものが、糖尿病の発症に対する相関の度合いが高い項目である。 The items with a large absolute value of the partial regression coefficient of each item of the health condition information obtained by the above analysis are items having a high degree of correlation with the onset of diabetes.
　続いて、発症可能性数値化部１０１は、健康状態情報の項目から、相関の度合いの高い項目を一定数抽出する（ＳＴＥＰ４０３３）。発症可能性数値化部１０１は、例えば、相関の度合いの高い順に１０項目を選び出すなどの方法により抽出する。 Subsequently, the onset probability digitization unit 101 extracts a certain number of items having a high degree of correlation from the items of the health condition information (STEP 4033). The onset probability quantification unit 101 extracts, for example, a method such as selecting 10 items in descending order of the degree of correlation.
　そして、発症可能性数値化部１０１は、例えば図３Ｂに示されるような形式にて、対象の疾病名を疾病名として、抽出された項目名を各疾病の発症可能性と相関のある項目としてそれぞれ発症可能性相関項目記憶部１１２に記憶する（ＳＴＥＰ４０３４）。 Then, the onset probability quantification unit 101 uses the extracted disease name as an item correlated with the onset probability of each disease in the format as shown in FIG. 3B, for example. Each is stored in the onset likelihood correlation item storage unit 112 (STEP 4034).
　発症可能性数値化部１０１は、発症可能性相関項目記憶部１１２に記憶されているすべての疾病についてのこれらの処理（ＳＴＥＰ４０３１～４０３４）を終えたときにはループＬ３を抜け、発症可能性相関項目抽出処理を終了する。 When the onset probability quantification unit 101 finishes these processes (STEPs 4031 to 4034) for all the diseases stored in the onset probability correlation item storage unit 112, it exits the loop L3 and extracts the onset probability correlation item. End the process.
　（発症可能性算出モデル準備処理）
　発症可能性算出モデル準備処理（ＳＴＥＰ４０４）は、任意の時点における任意の人の健康状態情報を入力として当該人の疾病ごとの発症可能性の数値を出力とするモデルである発症可能性算出モデルに用いる定数及び健康状態情報の各項目の係数を導出する処理である。処理内容は、図７Ｂに示される通りであり、発症可能性相関項目記憶部１１２に記憶されている疾病ごと（ループＬ４）に行われる。 (Probability calculation model preparation process)
The onset probability calculation model preparation process (STEP 404) is an onset probability calculation model which is a model that receives the health status information of an arbitrary person at an arbitrary time point and outputs a numerical value of the onset probability for each person's disease. This is a process for deriving coefficients for each item of constants and health condition information to be used. The processing content is as shown in FIG. 7B and is performed for each disease (loop L4) stored in the onset possibility correlation item storage unit 112.
　まず、発症可能性数値化部１０１は、発症可能性相関項目記憶部１１２を参照し、対象の疾病の発症可能性と相関のある健康状態情報の項目名を取得する（ＳＴＥＰ４０４１）。 First, the onset probability quantification unit 101 refers to the onset probability correlation item storage unit 112 and acquires the item name of the health condition information correlated with the onset probability of the target disease (STEP 4041).
　すなわち例えば、対象の疾病が糖尿病である場合、発症可能性数値化部１０１は、発症可能性と相関のある項目として記憶されたものが、血糖値、腹囲、飲酒習慣、ＨｂＡ１ｃ、運動習慣、喫煙習慣、身長、体重、年齢及び性別であることを認識する。 That is, for example, when the target disease is diabetes, the onset probability quantification unit 101 stores the blood sugar level, waist circumference, drinking habits, HbA1c, exercise habits, smoking, as items that are correlated with the onset possibility. Recognize habits, height, weight, age and gender.
　次に、発症可能性数値化部１０１は、健康状態記憶部１１１に記憶されている複数の人の最新の健康状態情報の各項目のうち、ＳＴＥＰ４０４１にて取得された対象の疾病の発症可能性と相関のある健康状態情報の各項目の値を抽出する（ＳＴＥＰ４０４２）。 Next, the onset probability quantification unit 101 includes the possibility of onset of the target disease acquired in STEP 4041 among the items of the latest health state information of a plurality of persons stored in the health state storage unit 111. The value of each item of the health condition information correlated with is extracted (STEP 4042).
　次に、発症可能性数値化部１０１は、対象の疾病の発症可能性算出モデルに用いる定数及び健康状態情報の各項目の係数を導出する（ＳＴＥＰ４０４３）。このモデルは、発症可能性相関項目抽出処理（ＳＴＥＰ４０３）におけるモデルと同種であり、例えばロジスティック回帰モデルである。 Next, the onset probability quantification unit 101 derives constants used in the onset probability calculation model of the target disease and coefficients of each item of the health condition information (STEP 4043). This model is the same as the model in the onset possibility correlation item extraction process (STEP 403), and is, for example, a logistic regression model.
　すなわち例えば、糖尿病の発症可能性に対する発症可能性算出モデルを生成又は更新する場合においては、発症可能性数値化部１０１は、対象の人の糖尿病の発症確率を目的変数とし、ＳＴＥＰ４０４２にて抽出した対象の疾病の発症可能性と相関のある項目の値を説明変数としたモデルの、定数及び各項目の偏回帰係数をロジスティック回帰分析により求める。 That is, for example, in the case of generating or updating the onset probability calculation model for the onset possibility of diabetes, the onset possibility quantification unit 101 uses the target person's diabetes onset probability as an objective variable and is extracted in STEP 4042 A constant and a partial regression coefficient of each item of a model having the value of an item correlated with the possibility of developing the target disease as an explanatory variable are obtained by logistic regression analysis.
　発症可能性数値化部１０１は、このような処理により求めた定数、各項目の偏回帰係数をロジスティック回帰分析の式に当てはめることにより、対象の疾病の発症可能性算出モデルを生成する（ＳＴＥＰ４０４４）。 The onset probability quantification unit 101 generates a model for calculating the onset probability of the target disease by applying the constants obtained by such processing and the partial regression coefficient of each item to the logistic regression analysis formula (STEP 4044). .
　発症可能性数値化部１０１は、生成又は更新した各疾病の発症可能性算出モデルを算出モデル記憶部１１３に記憶する（ＳＴＥＰ４０４５）。 The onset probability quantification unit 101 stores the generated or updated onset probability calculation model of each disease in the calculation model storage unit 113 (STEP 4045).
　なお、すでに当該疾病の発症可能性算出モデルが算出モデル記憶部１１３に記憶されている場合は、発症可能性数値化部１０１は、生成した発症可能性算出モデルで上書きをして記憶することで、健康状態の当該項目の発症可能性算出モデルを更新する。 In addition, when the onset probability calculation model of the disease is already stored in the calculation model storage unit 113, the onset probability digitizing unit 101 overwrites and stores the generated onset probability calculation model. Update the onset probability calculation model for the item of health status.
　発症可能性数値化部１０１は、発症可能性相関項目記憶部１１２に記憶されたすべての疾病についてこれらの処理（ＳＴＥＰ４０４１～ＳＴＥＰ４０４５）を終えたときにはループＬ４を抜け、発症可能性数値化モデル処理を終了する。 The onset probability quantification unit 101 exits the loop L4 when all the diseases stored in the onset probability correlation item storage unit 112 have been completed (STEP 4041 to STEP 4045), and performs the onset probability quantification model processing. finish.
（健康状態予測処理）
　健康状態予測処理（ＳＴＥＰ５０）は、各人４０の第２時点における予測される健康状態を示す情報を取得する処理である。当該処理の詳細については、図８及び図９を用いて説明する。 (Health condition prediction process)
The health state prediction process (STEP 50) is a process for acquiring information indicating the predicted health state of each person 40 at the second time point. Details of this processing will be described with reference to FIGS.
　処理内容は図８に示される通りであり、健康状態記憶部１１１に記憶されている健康状態情報に含まれるＩＤごと（ループＬ５）に行われる。また、ＳＴＥＰ５０３～５０６については、健康状態情報の項目ごと（ループＬ６）に行われる。 The processing content is as shown in FIG. 8 and is performed for each ID (loop L5) included in the health condition information stored in the health condition storage unit 111. Further, STEPs 503 to 506 are performed for each item of health condition information (loop L6).
　まず予測部１０２は、健康状態記憶部１１１から、対象のＩＤの１年前の健康状態情報Ｊ１及び最新の健康状態情報Ｊ２を取得する（ＳＴＥＰ５０１、図９）。 First, the prediction unit 102 acquires the health state information J1 and the latest health state information J2 of the target ID one year ago from the health state storage unit 111 (STEP 501, FIG. 9).
　次に、予測部１０２は、取得した最新の健康状態情報の各項目の値と１年前の健康状態情報の各項目の値との差分Ｊ１１を当該ＩＤについて計算する（ＳＴＥＰ５０２、図９）。 Next, the prediction unit 102 calculates a difference J11 between the value of each item of the acquired latest health condition information and the value of each item of the health condition information one year ago for the ID (STEP 502, FIG. 9).
　なお、例えば運動習慣や喫煙習慣のように、数値での測定がなされない項目については、予測部１０２は、最新の健康状態情報の値と１年前の健康状態情報の値との差分Ｊ１１を求めて続く処理において処理の対象に含めてもよいし、最新の健康状態情報の値と１年前の健康状態情報の値との差分Ｊ１１を求めずかつ続く処理において処理の対象に含めないこととしてもよい。 For items that are not measured numerically, such as exercise habits and smoking habits, for example, the prediction unit 102 calculates the difference J11 between the value of the latest health condition information and the value of the health condition information one year ago. It may be included in the processing target in the subsequent processing, or the difference J11 between the value of the latest health state information and the health state information one year ago is not calculated, and is not included in the processing target in the subsequent processing. It is good.
　数値での測定がなされない項目について処理対象に含める場合は、予測部１０２は、例えば喫煙習慣であれば、習慣ありを１と、習慣なしを０とするなどのように、各項目の値を数値に置き換えて取り扱う。 When an item that is not measured numerically is included in the processing target, the prediction unit 102 sets the value of each item such as 1 for habit and 0 for no habit, for example, if it is a smoking habit. Replace with numeric values.
　次に、予測部１０２は、変化相関項目記憶部１１４を参照し、対象の健康状態情報の項目の変化の大きさと相関のある健康状態情報の項目を取得する（ＳＴＥＰ５０３）。 Next, the prediction unit 102 refers to the change correlation item storage unit 114, and acquires items of health state information correlated with the magnitude of change of the target health state information item (STEP 503).
　次に、予測部１０２は、対象の健康状態情報の項目について、対象のＩＤの最新の健康状態情報から、対象の健康状態情報の項目の変化の大きさと相関のある項目Ｊ１３を抽出する（ＳＴＥＰ５０４、図９）。 Next, the prediction unit 102 extracts an item J13 having a correlation with the magnitude of change in the item of the target health condition information from the latest health condition information of the target ID for the item of the target health condition information (STEP 504). , FIG. 9).
　すなわち例えば、対象の健康状態情報の項目が腹囲である場合、健康状態情報の項目の値の変化と相関のある項目として記憶されたものは喫煙習慣、運動習慣、飲酒習慣、年齢及び性別であるため、予測部１０２は、これらの項目の値を当該ＩＤの最新の健康状態情報から抽出する。 That is, for example, when the item of health condition information of the subject is abdominal circumference, what is stored as an item correlated with a change in the value of the health condition information item is smoking habits, exercise habits, drinking habits, age and sex Therefore, the prediction unit 102 extracts the values of these items from the latest health condition information of the ID.
　次に、予測部１０２は、予測モデル記憶部１１５を参照し、対象の健康状態情報の項目の第２時点情報取得モデルＦ１を取得する（ＳＴＥＰ５０５、図９）。 Next, the prediction unit 102 refers to the prediction model storage unit 115 and acquires the second time point information acquisition model F1 of the item of the target health condition information (STEP 505, FIG. 9).
　次に、予測部１０２は、対象のＩＤの対象の項目の最新の値と１年前の値との差分Ｊ１１、対象のＩＤの対象の項目の最新の値Ｊ１２及び対象の項目と相関のある各項目Ｊ１３の値に基づいて対象のＩＤの対象の項目の第２時点における値を算出する（ＳＴＥＰ５０６、図９）。 Next, the prediction unit 102 correlates with the difference J11 between the latest value of the target item of the target ID and the value one year ago, the latest value J12 of the target item of the target ID, and the target item. Based on the value of each item J13, the value at the second time point of the target item of the target ID is calculated (STEP 506, FIG. 9).
　予測部１０２は、ＳＴＥＰ５０５にて取得したモデルＦ１に、対象の項目の最新の値と１年前の値との差分Ｊ１１、最新の値Ｊ１２及び対象の項目と相関のある各項目Ｊ１３の値を入力することにより、対象の健康状態情報の項目の最新の値と第２時点における値との差分の予測値Ｖ１を算出する（図９）。 The prediction unit 102 adds the difference J11 between the latest value of the target item and the value one year ago, the latest value J12, and the value of each item J13 correlated with the target item to the model F1 acquired in STEP 505. By inputting, a predicted value V1 of a difference between the latest value of the item of the health condition information of the target and the value at the second time point is calculated (FIG. 9).
　さらに予測部１０２は、得られたＶ１と予測対象の項目の最新の値Ｊ１２とを足し合わせて、第２時点である１年後の予測対象の項目の予測値Ｖ２を取得する（図９）。 Further, the prediction unit 102 adds the obtained V1 and the latest value J12 of the item to be predicted, and obtains the predicted value V2 of the item to be predicted one year later as the second time point (FIG. 9). .
　予測部１０２は、対象の健康状態情報の項目の最新の値と第２時点における値との差分の予測値Ｖ１の例えば９５％信頼区間の値を求めることとしてもよい。 The prediction unit 102 may obtain, for example, a 95% confidence interval value of the predicted value V1 of the difference between the latest value of the target health condition information item and the value at the second time point.
　この場合、予測部１０２は、対象の健康状態情報の項目の最新の値と第２時点における値との差分の予測値Ｖ１及び対象の項目の第２時点の予測値Ｖ２として一定の幅のある値を取得することができる（図９）。 In this case, the prediction unit 102 has a certain range as the predicted value V1 of the difference between the latest value of the item of the target health condition information and the value at the second time point and the predicted value V2 of the target item at the second time point. The value can be acquired (FIG. 9).
　そして、予測部１０２は、ＳＴＥＰ５０６にて算出した対象のＩＤの対象の項目の第２時点における値を予測値情報記憶部１１６に記憶する（ＳＴＥＰ５０７）。 Then, the prediction unit 102 stores the value at the second time point of the target item of the target ID calculated in STEP 506 in the predicted value information storage unit 116 (STEP 507).
　予測部１０２は、健康状態情報のすべての項目についてＳＴＥＰ５０３～５０７の処理を終えた時にはループＬ６を抜け、健康状態記憶部１１１に記憶されている健康状態情報に含まれるすべてのＩＤについてＳＴＥＰ５０１～５０７の処理を終えた時にはループＬ５を抜け、健康状態予測処理を終了する。 When the prediction unit 102 finishes the processing of STEPs 503 to 507 for all items of the health condition information, it exits the loop L6, and STEPs 501 to 507 for all IDs included in the health condition information stored in the health condition storage unit 111. When the above process is finished, the process exits the loop L5 and ends the health state prediction process.
（発症可能性数値化処理）
　発症可能性数値化処理（ＳＴＥＰ６０）は、各人４０の最新及び第２時点の疾病の発症可能性を数値化する処理である。当該処理の詳細については、図１０～１２を用いて説明する。 (Probability quantification process)
The onset possibility quantification process (STEP 60) is a process for quantifying the latest onset and second onset illness of each person 40. Details of this processing will be described with reference to FIGS.
　なお、発症可能性数値化部１０１は、発症可能性数値化処理による各人４０の疾病ごとの発症可能性の数値化を任意の頻度により繰り返し行う。 The onset probability digitization unit 101 repeatedly quantifies the onset probability for each disease of each person 40 by the onset probability digitization process at an arbitrary frequency.
　処理内容は図１０に示される通りであり、健康状態記憶部１１１に記憶されている健康状態情報に含まれるＩＤごと（ループＬ７）に行われる。また、ＳＴＥＰ６０２～６０９については、発症可能性相関項目記憶部１１２に記憶されている疾病ごと（ループＬ８）に行われる。 The processing content is as shown in FIG. 10 and is performed for each ID (loop L7) included in the health condition information stored in the health condition storage unit 111. Further, STEPs 602 to 609 are performed for each disease (loop L8) stored in the onset possibility correlation item storage unit 112.
　まず、発症可能性数値化部１０１は、健康状態記憶部１１１に記憶されている対象のＩＤの最新の健康状態情報、予測値情報記憶部１１６に記憶されている対象のＩＤの第２時点の健康状態情報、発症可能性数値情報記憶部１１７に記憶されている対象のＩＤの１年前の発症可能性数値情報を取得する（ＳＴＥＰ６０１）。 First, the symptom probability quantification unit 101 stores the latest health state information of the target ID stored in the health state storage unit 111 and the second time point of the target ID stored in the predicted value information storage unit 116. The onset probability numerical value information of the subject ID stored in the health state information and onset probability numerical information storage unit 117 one year ago is acquired (STEP 601).
　次に、発症可能性数値化部１０１は、発症可能性相関項目記憶部を参照し、対象の疾病の発症可能性と相関のある健康状態情報の項目を取得する（ＳＴＥＰ６０２）。 Next, the onset probability digitizing unit 101 refers to the onset probability correlation item storage unit, and acquires items of health state information correlated with the onset possibility of the target disease (STEP 602).
　続いて、発症可能性数値化部１０１は、疾病ごとの発症可能性と相関のある項目の値を対象のＩＤの最新の健康状態情報から抽出する（ＳＴＥＰ６０３）。 Subsequently, the onset probability digitizing unit 101 extracts the value of the item correlated with the onset probability for each disease from the latest health state information of the target ID (STEP 603).
　以下、対象の疾病が糖尿病である場合を例として本処理を説明する。 Hereinafter, this processing will be described by taking as an example the case where the target disease is diabetes.
　すなわち、疾病の発症可能性と相関のある項目として記憶されたものは血糖値、腹囲、飲酒習慣、ＨｂＡ１ｃ、運動習慣、喫煙習慣、身長、体重、年齢、性別であるため、発症可能性数値化部１０１は、これらの項目の値を対象のＩＤの最新の健康状態情報Ｊ３から抽出する（図１１）。 That is, since the items stored as items correlated with the possibility of developing the disease are blood glucose level, waist circumference, drinking habits, HbA1c, exercise habits, smoking habits, height, weight, age, and sex, The unit 101 extracts the values of these items from the latest health state information J3 of the target ID (FIG. 11).
　次に、発症可能性数値化部１０１は、算出モデル記憶部１１３を参照し、糖尿病の発症可能性算出モデルＦ２を取得する（ＳＴＥＰ６０４、図１１）。 Next, the onset probability quantification unit 101 refers to the calculation model storage unit 113 and acquires the onset probability calculation model F2 of diabetes (STEP 604, FIG. 11).
　次に、発症可能性数値化部１０１は、ＳＴＥＰ６０３にて抽出した糖尿病の発症可能性と相関のある各項目Ｊ３１の値に基づいて、対象のＩＤの対象の疾病の最新の発症可能性を数値化する（ＳＴＥＰ６０５、図１１）。 Next, the onset possibility quantification unit 101 numerically calculates the latest onset possibility of the target disease of the target ID based on the value of each item J31 correlated with the onset possibility of diabetes extracted in STEP 603. (STEP 605, FIG. 11).
　発症可能性数値化部１０１は、ＳＴＥＰ６０４にて取得したモデルＦ２に、ＳＴＥＰ６０３にて抽出した糖尿病の発症可能性と相関のある各項目Ｊ３１の値を入力し、対象の疾病の最新の発症可能性の数値Ｖ３を算出する（図１１）。 The onset possibility quantification unit 101 inputs the value of each item J31 correlated with the onset possibility of diabetes extracted in STEP 603 to the model F2 acquired in STEP 604, and the latest onset possibility of the target disease. The numerical value V3 is calculated (FIG. 11).
　なお、例えば対象の疾病が脂質異常症である場合は、発症可能性数値化部１０１は、脂質異常症の発症可能性と相関のある項目Ｊ３２に基づき、脂質異常症用の発症可能性算出モデルＦ３を用いて最新の発症可能性を数値化する（図１１）。 For example, when the target disease is dyslipidemia, the onset probability quantification unit 101 calculates the onset probability calculation model for dyslipidemia based on the item J32 correlated with the onset possibility of dyslipidemia. The latest possibility of onset is quantified using F3 (FIG. 11).
　次に、発症可能性数値化部１０１は、対象の疾病の発症可能性と相関のある項目の値を対象のＩＤの第２時点における予測される健康状態を示す情報から抽出する（ＳＴＥＰ６０６）。 Next, the onset probability digitization unit 101 extracts the value of the item correlated with the onset possibility of the target disease from the information indicating the predicted health state at the second time point of the target ID (STEP 606).
　すなわち、糖尿病の発症可能性と相関のある項目として記憶された健康状態情報の項目は血糖値、腹囲、飲酒習慣、ＨｂＡ１ｃ、運動習慣、喫煙習慣、身長、体重、年齢、性別であるため、発症可能性数値化部１０１は、これらの項目の値をＳＴＥＰ６０１にて取得した対象のＩＤの第２時点における健康状態情報Ｊ４から抽出する（図１２）。 That is, since the items of health condition information stored as items correlated with the possibility of developing diabetes are blood glucose level, waist circumference, drinking habits, HbA1c, exercise habits, smoking habits, height, weight, age, sex, The possibility digitizing unit 101 extracts the values of these items from the health state information J4 at the second time point of the target ID acquired in STEP 601 (FIG. 12).
　なお、健康状態予測処理（ＳＴＥＰ５０）において、予測部１０２が、運動習慣や喫煙習慣などの数値での測定がなされない項目について処理の対象に含めなかった場合は、発症可能性数値化部１０１は、これらの情報については、対象のＩＤの最新の健康状態情報Ｊ３から抽出する（図１２）。 In addition, in the health state prediction process (STEP 50), when the prediction unit 102 does not include items that are not measured with numerical values such as exercise habits and smoking habits, the likelihood of occurrence quantification unit 101 These pieces of information are extracted from the latest health condition information J3 of the target ID (FIG. 12).
　次に、発症可能性数値化部１０１は、抽出した糖尿病の発症可能性と相関のある項目Ｊ４１に基づいて、対象のＩＤの対象の疾病の第２時点の発症可能性を数値化する（ＳＴＥＰ６０７）。 Next, the onset possibility quantification unit 101 quantifies the onset possibility at the second time point of the target disease of the target ID based on the item J41 correlated with the extracted onset possibility of diabetes (STEP 607). ).
　この処理を実行する際には、ＳＴＥＰ６０４で取得した発症可能性算出モデルＦ２を再度用いる。 When executing this process, the onset possibility calculation model F2 acquired in STEP 604 is used again.
　当該モデルに、ＳＴＥＰ６０５にて抽出した糖尿病の発症可能性と相関のある各項目Ｊ４１の値を入力し、対象のＩＤの対象の疾病の第２時点の発症可能性の数値Ｖ４を算出する（図１２）。 The value of each item J41 correlated with the possibility of the onset of diabetes extracted in STEP 605 is input to the model, and the numerical value V4 of the onset probability at the second time point of the target disease of the target ID is calculated (FIG. 12).
　なお、健康状態予測処理のＳＴＥＰ５０６において、予測部１０２が対象の項目の第２時点の予測値Ｖ２として一定の幅のある値を取得し、予測値情報記憶部１１６に記憶している場合には、発症可能性数値化部１０１は、対象の疾病の第２時点の発症可能性の数値Ｖ４として、幅のある値を算出することができる。（図１２）
　この場合、発症可能性数値化部１０１は、例えば疾病の発症可能性と相関のある各項目Ｊ４１の値のそれぞれの最大値、中央値及び最小値を取得し、最大値だけの組み合わせ、最小値だけの組み合わせ、最大値、中央値及び最小値による任意の組み合わせなどの様々なパターンで発症可能性算出モデルＦ２に入力して対象の疾病の第２時点の発症可能性の数値Ｖ４を算出し、得られた算出結果の値のうち最大の値を当該疾病の発症可能性の数値Ｖ４の最大値、得られた算出結果の値のうち最小の値を当該疾病の発症可能性の数値Ｖ４の最小値として算出する（図１２）。 In STEP 506 of the health state prediction process, when the prediction unit 102 acquires a value with a certain width as the predicted value V2 at the second time point of the target item and stores it in the predicted value information storage unit 116. The onset probability digitization unit 101 can calculate a wide value as the onset probability value V4 at the second time point of the target disease. (Fig. 12)
In this case, the onset probability quantification unit 101 acquires, for example, the maximum value, the median value, and the minimum value of each item J41 correlated with the onset probability of the disease, and the combination of only the maximum value, the minimum value Only the combination, the maximum value, the median, and any combination of the minimum value, etc., are input to the onset probability calculation model F2 to calculate the numerical value V4 of the onset probability at the second time point of the target disease, Among the obtained calculation result values, the maximum value is the maximum value of the disease onset probability value V4, and the obtained calculation result value is the minimum value of the disease onset probability value V4. Calculated as a value (FIG. 12).
　また、例えば対象の疾病が脂質異常症である場合は、発症可能性数値化部１０１は、脂質異常症の発症可能性と相関のある項目Ｊ４２に基づき、脂質異常症用の発症可能性算出モデルＦ３を用いて最新の発症可能性を数値化する（図１２）。 For example, when the target disease is dyslipidemia, the onset probability quantification unit 101 calculates the onset probability calculation model for dyslipidemia based on the item J42 correlated with the onset possibility of dyslipidemia. The latest possibility of onset is quantified using F3 (FIG. 12).
　続いて、発症可能性数値化部１０１は、対象のＩＤの最新、第２時点及び１年前の疾病の発症可能性の数値に応じた発症可能性の種別を設定する（ＳＴＥＰ６０８）。 Subsequently, the onset probability digitization unit 101 sets the type of onset possibility according to the latest ID of the target ID, the second time point, and the onset probability of the disease one year ago (STEP 608).
　発症可能性数値化部１０１は、例えば変換テーブル記憶部１１９に記憶された図１４Ａに示されるような変換テーブルを上から順に参照することで発症可能性の種別を判定し、設定する。 The onset probability digitizing unit 101 determines and sets the type of onset possibility by referring to the conversion table as shown in FIG. 14A stored in the conversion table storage unit 119 in order from the top, for example.
　すなわち例えばある疾病の発症可能性の数値が１５１～２００であれば、発症可能性数値化部１０１は、発症可能性の種別を４と設定する。 That is, for example, if the numerical value of the onset probability of a certain disease is 151 to 200, the onset probability digitizing unit 101 sets the type of the onset probability to 4.
　あるいは例えば、ある疾病の発症可能性の数値が発症可能性の種別の４及び３には該当せず、発症可能性の数値はいずれも１００以下だが前回と比して５０以上上昇した疾病が１つ以上ある場合は、発症可能性数値化部１０１は、発症可能性の種別を２と設定する。 Or, for example, the numerical value of the probability of developing a certain disease does not correspond to the types 4 and 3 of the probability of developing, and the numerical value of the probability of developing is 100 or less, but a disease that has increased by 50 or more compared to the previous time is 1 If there is more than one, the onset probability quantification unit 101 sets the type of onset possibility to 2.
　発症可能性数値化部１０１は、発症可能性の数値が前回と比して５０以上上昇した疾病が１つ以上あるか否かの判定を、例えばＳＴＥＰ６０５にて算出した対象のＩＤの各疾病の最新の発症可能性の数値とＳＴＥＰ６０１にて取得した当該ＩＤの当該疾病の１年前の発症可能性数値情報の値とを比較し、発症可能性の数値が５０以上上昇した疾病が１つ以上あるか否かにより行う。 The onset probability quantification unit 101 determines whether there is one or more diseases whose onset probability has increased by 50 or more compared to the previous time, for example, for each disease of the target ID calculated in STEP 605 Compare the latest numerical value of the likelihood of onset and the value of the numerical value of the probability of onset one year before the disease with the ID acquired in STEP601, and one or more diseases whose numerical value of the probability of onset has increased by 50 or more Depending on whether or not there is.
　あるいは例えば、発症可能性数値化部１０１は、ある疾病の発症可能性の発症可能性の種別の４～２の何れにも該当しないが、第２時点の発症可能性の数値が１００以上であれば、発症可能性の種別を１と設定する。 Or, for example, the onset probability quantification unit 101 does not correspond to any of the 4 to 2 types of onset possibility of the onset possibility of a certain disease, but the onset possibility value at the second time point is 100 or more. For example, the type of onset possibility is set to 1.
　そして、発症可能性数値化部１０１は、対象のＩＤの最新の疾病の発症可能性数値情報及び発症可能性の種別を発症可能性数値情報記憶部１１７に、対象のＩＤの第２時点の疾病の発症可能性数値情報を予測値情報記憶部１１６に、それぞれ記憶する（ＳＴＥＰ６０９）。 Then, the onset probability digitizing unit 101 stores the latest onset probability numerical information on the target ID and the type of onset possibility in the onset probability numerical information storage unit 117, and the disease at the second time point of the target ID. Are stored in the predicted value information storage unit 116 (STEP 609).
　発症可能性数値化部１０１は、発症可能性相関項目記憶部１１２に記憶されているすべての疾病についてＳＴＥＰ６０２～６０９の処理を終えた時にはループＬ８を抜け、健康状態記憶部１１１に記憶されている健康状態情報に含まれるすべてのＩＤについてＳＴＥＰ６０１～６０９の処理を終えたときに、発症可能性数値化処理を終了する。 The onset probability quantification unit 101 exits the loop L8 when all the diseases stored in the onset probability correlation item storage unit 112 have been processed in STEPs 602 to 609, and is stored in the health state storage unit 111. When the processing of STEPs 601 to 609 is completed for all the IDs included in the health status information, the onset probability quantification processing is terminated.
　（注意情報生成処理）
　注意情報生成処理（ＳＴＥＰ８０）は、各人４０の疾病の発症可能性の数値に応じて疾病の発症可能性に関する情報を含む疾病注意情報を生成する処理である。処理内容は、図１３に示される通りであり、健康状態記憶部１１１に記憶されている健康状態情報に含まれるＩＤごと（ループＬ９）に行われる。 (Caution information generation process)
The attention information generation process (STEP 80) is a process for generating disease attention information including information on the possibility of the onset of the disease according to the numerical value of the onset of the disease of each person 40. The processing content is as shown in FIG. 13 and is performed for each ID (loop L9) included in the health condition information stored in the health condition storage unit 111.
　なお、注意情報生成部１０３による疾病注意情報の生成は任意の頻度により繰り返し行われる。 Note that the generation of the disease attention information by the attention information generation unit 103 is repeatedly performed at an arbitrary frequency.
　まず、注意情報生成部１０３は、対象のＩＤの最新の発症可能性の種別を取得し（ＳＴＥＰ８０１）、当該発症可能性の種別に応じた疾病注意情報の生成頻度を認識する（ＳＴＥＰ８０２）。 First, the attention information generation unit 103 acquires the latest type of possibility of onset of the target ID (STEP 801), and recognizes the generation frequency of the disease attention information according to the type of the possibility of onset (STEP 802).
　注意情報生成部１０３は、例えば変換テーブル記憶部１１９に記憶された図１４Ｂに示されるような変換テーブル参照して、対象のＩＤの最新の疾病の発症可能性の種別に応じた疾病注意情報の生成頻度を認識する。 The attention information generation unit 103 refers to the conversion table as shown in FIG. 14B stored in the conversion table storage unit 119, for example, and stores the disease attention information according to the type of the latest disease onset possibility of the target ID. Recognize the generation frequency.
　すなわち例えば注意情報生成部１０３は、対象のＩＤの疾病の発症可能性の種別の最大値が４である場合に、対象のＩＤの疾病注意情報の生成頻度を毎月と設定する。 That is, for example, the attention information generation unit 103 sets the generation frequency of the disease attention information of the target ID as monthly when the maximum value of the type of the possibility of developing the disease of the target ID is 4.
　そして、注意情報生成部１０３は、対象のＩＤについて疾病注意情報を前回生成してから、疾病注意情報の生成頻度以上の時間を経過しているか否かを判定する（ＳＴＥＰ８０３）。 Then, the attention information generation unit 103 determines whether or not a time equal to or greater than the generation frequency of the disease attention information has elapsed since the disease attention information was previously generated for the target ID (STEP 803).
　注意情報生成部１０３は、当該判定を、図３Ｆに示されるような注意情報最終年月日情報を参照し、各対象のＩＤの注意情報最終年月日と、注意情報当該処理を行っている日付とを比較して、生成頻度以上の時間を経過しているか否かを判定することにより行う。 The attention information generation unit 103 refers to the attention information last date information as shown in FIG. 3F for the determination, and performs the attention information last date of each target ID and the attention information concerned processing. This is done by comparing the date and determining whether or not a time equal to or greater than the generation frequency has elapsed.
　例えば、発症可能性の数値はいずれも１００以下だが前回と比して５０以上上昇した疾病が１つ以上ある場合は発症可能性の種別が２に設定されているので、疾病注意情報の生成頻度は半年に１回であるので、注意情報生成部１０３は、当該ＩＤについての注意情報最終年月日から半年を経過していれば、生成頻度以上の時間を経過していると判定し、経過していなければ生成頻度以上の時間を経過していないと判定する。 For example, if there is one or more illnesses that have a probability of onset of 100 or less but increased by 50 or more compared to the previous time, the symptom probability type is set to 2, so the frequency of occurrence of disease caution information Is once in half a year, the attention information generation unit 103 determines that the time more than the generation frequency has passed if half a year has passed since the last date of the attention information about the ID. If not, it is determined that the time exceeding the generation frequency has not elapsed.
　当該判定が否定的（ＳＴＥＰ８０３：ＮＯ）であれば、当該ＩＤについては、疾病注意情報を生成する頻度の条件を満たしていないので、注意情報生成部１０３は、ＳＴＥＰ８０４～８０５を実行せずに、次の人に処理の対象者を移したうえでＳＴＥＰ８０１以降の処理を繰り返す。 If the determination is negative (STEP 803: NO), since the condition for the frequency of generating disease attention information is not satisfied for the ID, the attention information generation unit 103 does not execute STEPs 804 to 805, After transferring the person to be processed to the next person, the processing after STEP 801 is repeated.
　一方当該判定が肯定的（ＳＴＥＰ８０３：ＹＥＳ）であれば、当該ＩＤについては、疾病注意情報を生成する頻度の条件を満たしているので、注意情報生成部１０３は、疾病注意情報を作成するため、対象のＩＤの最新の健康状態情報、１年前の健康状態情報及び第２時点の健康状態情報を取得する（ＳＴＥＰ８０４）。 On the other hand, if the determination is affirmative (STEP 803: YES), since the condition for the frequency of generating the disease attention information is satisfied for the ID, the attention information generation unit 103 creates the disease attention information. The latest health condition information of the target ID, the health condition information one year ago, and the health condition information at the second time point are acquired (STEP 804).
　そして、注意情報生成部１０３は、対象のＩＤの疾病の最新及び第２時点のいずれか一方又は両方の発症可能性の数値に応じて疾病注意情報を生成する（ＳＴＥＰ８０５）。 Then, the attention information generation unit 103 generates disease attention information according to the numerical value of the possibility of onset of one or both of the latest disease of the target ID and the second time point (STEP 805).
　注意情報生成部１０３は、例えば変換テーブル記憶部１１９に記憶された図１４Ｃに示されるような変換テーブルを参照して、改善アドバイス情報３２２を生成する。 The attention information generation unit 103 generates the improvement advice information 322 with reference to the conversion table as illustrated in FIG. 14C stored in the conversion table storage unit 119, for example.
　例えば注意情報生成部１０３は、糖尿病の発症可能性の種別が４である場合に、「肥満を解消する、食事量を減らす、動物性脂肪や糖質（特に清涼飲料水）の過度な摂取を控える…」などの疾病と発症可能性の種別に対応する改善アドバイスを取得することにより改善アドバイス情報３２２を生成する。 For example, when the type of diabetes occurrence possibility is 4, the attention information generation unit 103 may “relieve obesity, reduce the amount of meals, excessive intake of animal fats and carbohydrates (especially soft drinks)”. The improvement advice information 322 is generated by acquiring the improvement advice corresponding to the type of the disease and the possibility of onset such as “Reserve ...”.
　なお、注意情報生成部１０３は、疾病の発症可能性の数値が一定値以上である疾病についてのみ前記改善アドバイス情報を生成するように構成されてもよい。 Note that the attention information generation unit 103 may be configured to generate the improvement advice information only for a disease for which the numerical value of the possibility of developing the disease is a certain value or more.
　注意情報生成部１０３は、例えば変換テーブル記憶部１１９に記憶された図１４Ｄに示されるような変換テーブルを参照して合併症に関する情報３２３を生成する。 The attention information generation unit 103 generates complication information 323 with reference to a conversion table as illustrated in FIG. 14D stored in the conversion table storage unit 119, for example.
　例えば注意情報生成部１０３は、糖尿病を発症している又は発症する可能性のある者に対しては、「心筋梗塞、脳卒中、腎不全」などの合併症に関する情報を取得することにより合併症に関する情報３２３を生成する。 For example, the attention information generation unit 103 obtains information on complications such as “myocardial infarction, stroke, renal failure” for those who have or may develop diabetes. Information 323 is generated.
　注意情報生成部１０３は、疾病ごとの発症可能性の数値を表すグラフ３２４を、疾病ごとの発症可能性の数値が高い順に並べるように構成されてもよい。 The attention information generation unit 103 may be configured to arrange the graph 324 representing the numerical value of the onset probability for each disease in descending order of the numerical value of the onset probability for each disease.
　注意情報生成部１０３は、例えば同じ性別かつ同じ年代であるＩＤごとに複数の人の全部又は一部により構成されるグループを作成し、当該グループ内の各ＩＤの疾病ごとの発症可能性の順位を計算することにより各人４０の疾病ごとの発症可能性の順位に関する情報３２５を生成する。 The attention information generation unit 103 creates a group composed of all or part of a plurality of people for each ID having the same gender and the same age, for example, and ranks of the onset probability for each disease of each ID in the group Is calculated to generate information 325 relating to the ranking of the onset probability of each person 40 for each disease.
　注意情報生成部１０３は、当該グループ内の実順位を計算し、例えば１９７５人中の５０位などと表してもよいし、あるいは、当該グループ内の順位を１００人中の順位に置き換えて計算し、小数点以下の数値がある場合は四捨五入し、１００人中３位などと表してもよい。 The attention information generation unit 103 may calculate the actual rank within the group and may represent, for example, the 50th place in 1975, or the rank within the group may be replaced with the rank of the 100 person. If there are numbers after the decimal point, they may be rounded off and expressed as 3rd place out of 100 people.
　あるいは、例えば端末３０から各人４０の勤務先名、勤務先の業種、職種などの情報を疾病注意情報提供支援サーバ１０が取り込み、注意情報生成部１０３は、それらの情報に基づいて複数の人の全部又は一部により構成されるグループを作成することとしてもよい。 Alternatively, for example, the illness caution information provision support server 10 captures information such as the work name of each person 40 from the terminal 30, the business type of the work, and the occupation type, and the caution information generation unit 103 uses a plurality of people based on the information. It is good also as creating the group comprised by all or one part.
　なお、注意情報生成部１０３は、改善アドバイス情報３２２、合併症に関する情報３２３、複数の人の全部又は一部により構成されるグループにおける各ＩＤの疾病ごとの発症可能性の順位に関する情報３２５などの情報を、各ＩＤの疾病ごとの最新の発症可能性の数値だけでなく、疾病ごとの第２時点における発症可能性の数値に基づいて生成することとしてもよい。 Note that the attention information generation unit 103 includes improvement advice information 322, information 323 relating to complications, information 325 relating to the ranking of the onset probability for each disease of each ID in a group composed of all or part of a plurality of persons, and the like. The information may be generated based on not only the latest numerical value of the onset possibility for each disease of each ID but also the numerical value of the onset possibility at the second time point for each disease.
　次に、注意情報生成部１０３は、疾病注意情報のＩＤごとの生成年月日を履歴記憶部１１８に記憶する（ＳＴＥＰ８０６）。 Next, the attention information generation unit 103 stores the generation date for each ID of the disease attention information in the history storage unit 118 (STEP 806).
　注意情報生成部１０３は、健康状態記憶部１１１に記憶されている健康状態情報に含まれるすべてのＩＤについてこれらの処理（ＳＴＥＰ８０１～８０６）が終わったときにはループＬ９を抜け、注意情報生成処理を終了する。 The attention information generation unit 103 exits the loop L9 when these processes (STEPs 801 to 806) are completed for all IDs included in the health state information stored in the health state storage unit 111, and ends the attention information generation process. To do.
　（他の実施形態）
　以上、本発明の実施形態の一例について説明したが、本発明の実施形態はこれに限定されない。 (Other embodiments)
The example of the embodiment of the present invention has been described above, but the embodiment of the present invention is not limited to this.
　たとえば、発症可能性数値化部１０１が各疾病の発症可能性を数値化する際に用いるモデル又は予測部１０２が第１時点より後の第２時点における対象の人の予測される健康状態を示す情報を取得する際に用いるモデルにかえて、ディープラーニングやサポートベクターマシン等の識別機を用いてもよい。 For example, the model or prediction unit 102 used when the onset probability quantification unit 101 quantifies the onset probability of each disease indicates the predicted health state of the target person at a second time point after the first time point. Instead of a model used when acquiring information, an identification machine such as deep learning or a support vector machine may be used.
１０…疾病注意情報提供支援サーバ、１００…サーバ制御部、１０１…発症可能性数値化部、１０２…予測部、１０３…注意情報生成部、１０４…注意情報送信部、１１０…サーバ記憶部、１１１…健康状態記憶部、１１２…発症可能性相関項目記憶部、１１３…算出モデル記憶部、１１４…変化相関項目記憶部、１１５…予測モデル記憶部、１１６…予測値情報記憶部、１１７…発症可能性数値情報記憶部、１１８…履歴記憶部、１１９…変換テーブル記憶部、２０…情報通信網、３０…端末、４０…各人。
  DESCRIPTION OF SYMBOLS 10 ... Disease attention information provision support server, 100 ... Server control part, 101 ... Probability digitization part, 102 ... Prediction part, 103 ... Caution information generation part, 104 ... Caution information transmission part, 110 ... Server memory part, 111 ... health condition storage unit, 112 ... onset possibility correlation item storage unit, 113 ... calculation model storage unit, 114 ... change correlation item storage unit, 115 ... prediction model storage unit, 116 ... prediction value information storage unit, 117 ... onset possibility Sex numerical value information storage unit, 118 ... history storage unit, 119 ... conversion table storage unit, 20 ... information communication network, 30 ... terminal, 40 ... each person.

Claims (12)

  1. 　１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶する健康状態記憶部と、
    　記憶した前記健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する発症可能性数値化部と、
    　前記発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部と、
    　任意の時点における任意の人の健康状態情報を含む情報を入力として当該時点より後の時点における該人の健康状態を示す情報を出力とするモデルに、第１時点における対象の人の健康状態情報を含む情報を入力し、該モデルから出力された情報により、該第１時点より後の第２時点における該対象の人の予測される健康状態を示す情報を取得する予測部を備え、
    　前記発症可能性数値化部は、前記予測部が取得した情報に基づいて各人の疾病ごとの前記第２時点における発症可能性を数値化し、
    　前記注意情報生成部は、疾病ごとの前記第２時点における発症可能性の数値に基づいて前記疾病注意情報を生成するように構成されていることを特徴とする疾病注意情報提供支援システム。     A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons;
    An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information;
    Attention information generation unit for generating disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or a plurality of diseases digitized by the onset probability digitization unit;
    Information on the health status of the subject at the first time is input to the model including the information including the health status information of any man at any time as an input and information indicating the health status of the man at a time after the time is output. Including a prediction unit that acquires information indicating a predicted health state of the target person at a second time point after the first time point according to the information output from the model,
    The onset probability quantification unit quantifies the onset probability at the second time point for each person's disease based on the information acquired by the prediction unit,
    The disease attention information providing support system, wherein the attention information generation unit is configured to generate the disease attention information based on a numerical value of the possibility of onset at the second time point for each disease.
  2. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記疾病注意情報は、疾病ごとの発症可能性の改善方法に関する情報である改善アドバイス情報を含み、
    　前記注意情報生成部は、疾病の発症可能性の数値が一定値以上である疾病についてのみ前記改善アドバイス情報を生成するように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The disease attention information includes improvement advice information that is information on a method for improving the possibility of onset for each disease,
    The disease attention information provision support system, wherein the attention information generation unit is configured to generate the improvement advice information only for a disease having a numerical value of a disease onset probability equal to or greater than a certain value.
  3. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記疾病注意情報は、疾病の発症可能性の数値に応じて各人が発症する可能性のある疾病に起因して発症する可能性のある合併症に関する情報を含むように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The disease attention information is configured to include information on complications that may occur due to a disease that each person may develop according to a numerical value of the possibility of developing the disease. A featured disease attention information provision support system.
  4. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記疾病注意情報は、各人の疾病ごとの発症可能性の数値を表すグラフを含み、該グラフは疾病ごとの発症可能性の数値が高い順に並べられていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The disease attention information includes a graph showing a numerical value of the probability of onset for each disease of each person, and the graph is arranged in descending order of the numerical value of the probability of onset for each disease Support system.
  5. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記疾病注意情報は、前記複数の人の全部又は一部により構成されるグループにおける各人の疾病ごとの発症可能性の順位に関する情報を含むように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The disease attention information is configured so that the disease attention information includes information related to a ranking of the probability of occurrence of each person's disease in a group composed of all or part of the plurality of persons. Offer support system.
  6. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記発症可能性数値化部は、前記健康状態情報から疾病ごとの発症可能性と相関のある健康状態情報を抽出し、相関のある健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化するように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The onset probability quantification unit extracts health state information correlated with the onset probability for each disease from the health state information, and determines the onset probability for each person's disease based on the correlated health state information. A disease attention information provision support system characterized by being configured to be digitized.
  7. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記予測部は、複数の人のそれぞれのある時点における健康状態情報を含む情報と、該複数の人のそれぞれの該時点の健康状態情報と該複数の人のそれぞれの該時点より前の時点の健康状態情報との間における変化の大きさを含む情報とを用いて、前記モデルを生成又は更新するように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The prediction unit includes information including health state information at each time point of the plurality of people, health state information at each time point of the plurality of people, and time points before the respective time points of the plurality of people. A disease caution information provision support system configured to generate or update the model using information including the magnitude of change between the health condition information.
  8. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記発症可能性数値化部は、各人の疾病ごとの発症可能性の数値化を繰り返し行い、前記注意情報生成部は、少なくとも該数値が一定以上変化した人向けの疾病注意情報を生成するように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The onset probability quantification unit repeatedly quantifies the onset probability of each person's disease, and the attention information generation unit generates disease attention information for a person whose value has changed at least a certain level. The disease attention information provision support system characterized by being comprised.
  9. 　請求項１に記載の疾病注意情報提供支援システムにおいて、
    　前記注意情報生成部は、各人の疾病ごとの発症可能性の数値に応じて疾病注意情報を生成する頻度を異なるものとするように構成されていることを特徴とする疾病注意情報提供支援システム。     In the disease attention information provision support system according to claim 1,
    The disease attention information providing support system, wherein the attention information generation unit is configured to vary the frequency of generating disease attention information according to a numerical value of the possibility of onset for each person's disease. .
  10. 　１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶する健康状態記憶部と、
    　記憶した前記健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する発症可能性数値化部と、
    　前記発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部とを備え、
    　前記疾病注意情報は、該疾病注意情報が提供される対象の人の疾病ごとの発症可能性について、前記複数の人の全部又は一部により構成されるグループにおける順位を示す情報を含むように構成されていることを特徴とする疾病注意情報提供支援システム。     A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons;
    An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information;
    An attention information generation unit that generates disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or more diseases digitized by the onset probability digitization unit,
    The disease attention information is configured to include information indicating a rank in a group composed of all or a part of the plurality of persons with respect to the possibility of onset for each disease of a person to whom the disease attention information is provided. Disease attention information provision support system characterized by being.
  11. 　１又は複数の人の健康状態に関する複数の項目の値を含む健康状態情報を記憶する健康状態記憶部と、
    　記憶した前記健康状態情報に基づいて各人の疾病ごとの発症可能性を数値化する発症可能性数値化部と、
    　前記発症可能性数値化部により数値化された1又は複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部とを備え、
    　前記発症可能性数値化部は、１又は複数の人の前記健康状態情報を分析することにより疾病ごとの発症可能性と相関のある該健康状態情報の項目を特定し、対象の人の健康状態情報から前記特定された疾病ごとの発症可能性と相関のある項目の値を抽出し、該抽出された項目の値に基づいて各人の疾病ごとの発症可能性を数値化するように構成されていることを特徴とする疾病注意情報提供支援システム。     A health condition storage unit that stores health condition information including values of a plurality of items relating to the health condition of one or more people;
    An onset probability quantification unit that quantifies the onset probability of each person's disease based on the stored health condition information;
    An attention information generation unit that generates disease attention information including information indicating the numerical value of the onset probability of at least one disease among one or more diseases digitized by the onset probability digitization unit,
    The onset probability quantification unit identifies the item of the health status information correlated with the onset probability for each disease by analyzing the health status information of one or a plurality of people, and the health status of the target person It is configured to extract a value of an item correlated with the onset possibility for each specified disease from the information, and to quantify the onset probability for each person's disease based on the value of the extracted item. A disease attention information provision support system characterized by
  12. 　１又は複数の人の健康状態に関する情報を含む健康状態情報を記憶する健康状態記憶部と、
    　記憶した前記健康状態情報に基づいて各人の複数の疾病それぞれの発症可能性を数値化する発症可能性数値化部と、
    　前記発症可能性数値化部により数値化された複数の疾病のうち少なくとも1つの疾病の発症可能性の数値を示す情報を含む疾病注意情報を生成する注意情報生成部とを備え、
    　前記注意情報生成部は、前記複数の疾病のうち発症可能性の数値が所定の条件を満たす疾病を認識し、該疾病に関連付けられて記憶された合併症を認識し、該合併症に関する情報を含む前記疾病注意情報を生成するように構成されていることを特徴とする疾病注意情報提供支援システム。
          A health condition storage unit that stores health condition information including information relating to the health condition of one or more persons;
    An onset probability quantification unit that quantifies the onset probability of each of a plurality of diseases of each person based on the stored health condition information;
    An attention information generation unit that generates disease attention information including information indicating a numerical value of the onset probability of at least one disease among a plurality of diseases digitized by the onset probability digitization unit,
    The attention information generating unit recognizes a disease whose numerical value of the probability of occurrence satisfies a predetermined condition among the plurality of diseases, recognizes a complication stored in association with the disease, and stores information on the complication A disease attention information provision support system, characterized in that the disease attention information is generated.
PCT/JP2018/047678 2018-03-01 2018-12-25 Disease-warning-information providing support system WO2019167406A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2018-037004 2018-03-01
JP2018037004A JP6436515B1 (en) 2018-03-01 2018-03-01 Disease attention information provision support system

Publications (1)

Publication Number Publication Date
WO2019167406A1 true WO2019167406A1 (en) 2019-09-06

Family

ID=64655900

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/047678 WO2019167406A1 (en) 2018-03-01 2018-12-25 Disease-warning-information providing support system

Country Status (2)

Country Link
JP (1) JP6436515B1 (en)
WO (1) WO2019167406A1 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7315165B2 (en) * 2019-06-10 2023-07-26 ＴＸＰ Ｍｅｄｉｃａｌ株式会社 Diagnosis support system
CN110277168A (en) * 2019-06-25 2019-09-24 珠海格力电器股份有限公司 A kind of method and apparatus of healthy early warning

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002157339A (en) * 2000-11-17 2002-05-31 Hoken Joho Service:Kk Health check and life habit improvement instruction system
JP2003216740A (en) * 2002-01-28 2003-07-31 Matsushita Electric Ind Co Ltd Method, system and program for health care and medium recorded with health care program
JP2006155411A (en) * 2004-11-30 2006-06-15 Houken Corp System, method, and program for constructing medical treatment prediction model, system, method, and program for verifying / offering medical treatment content, method of downloading health management data, and detachable storage medium
JP2017117469A (en) * 2015-12-22 2017-06-29 国立研究開発法人理化学研究所 Risk evaluation method, risk evaluation device, and risk evaluation program

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002157339A (en) * 2000-11-17 2002-05-31 Hoken Joho Service:Kk Health check and life habit improvement instruction system
JP2003216740A (en) * 2002-01-28 2003-07-31 Matsushita Electric Ind Co Ltd Method, system and program for health care and medium recorded with health care program
JP2006155411A (en) * 2004-11-30 2006-06-15 Houken Corp System, method, and program for constructing medical treatment prediction model, system, method, and program for verifying / offering medical treatment content, method of downloading health management data, and detachable storage medium
JP2017117469A (en) * 2015-12-22 2017-06-29 国立研究開発法人理化学研究所 Risk evaluation method, risk evaluation device, and risk evaluation program

Also Published As

Publication number Publication date
JP6436515B1 (en) 2018-12-12
JP2019153015A (en) 2019-09-12

Similar Documents

Publication Publication Date Title
US11664097B2 (en) Healthcare information technology system for predicting or preventing readmissions
US8949082B2 (en) Healthcare information technology system for predicting or preventing readmissions
JP5977898B1 (en) BEHAVIOR PREDICTION DEVICE, BEHAVIOR PREDICTION DEVICE CONTROL METHOD, AND BEHAVIOR PREDICTION DEVICE CONTROL PROGRAM
US20180107798A1 (en) Method for aiding a diagnosis, program and apparatus
US20060265253A1 (en) Patient data mining improvements
US20140095201A1 (en) Leveraging Public Health Data for Prediction and Prevention of Adverse Events
Baker et al. Continuous and automatic mortality risk prediction using vital signs in the intensive care unit: a hybrid neural network approach
CN108140175A (en) Labor management system, Method of labor management in road and labor management program
JP2018060529A (en) Method and apparatus of context-based patient similarity
Archenaa et al. Health recommender system using big data analytics
Damotte et al. Harnessing electronic medical records to advance research on multiple sclerosis
Zhang et al. Advanced diagnostic imaging utilization during emergency department visits in the United States: A predictive modeling study for emergency department triage
WO2019167406A1 (en) Disease-warning-information providing support system
CN111295716B (en) Health management assistance device, method, and program
Steurer et al. Clinical value of prognostic instruments to identify patients with an increased risk for osteoporotic fractures: systematic review
WO2016006042A1 (en) Data analysis device, control method for data analysis device, and control program for data analysis device
Gräßer et al. Application of recommender system methods for therapy decision support
JP6531241B1 (en) Efficiency improvement support system and medical efficiency improvement support method
GB2555381A (en) Method for aiding a diagnosis, program and apparatus
JP6436855B2 (en) Analysis system and analysis method
JP7082809B2 (en) Member health condition management system, member health condition management method, and member health condition management program
JP2021135930A (en) Analyzer, analytical method and analytical program
Kim et al. Disease risk prediction system using correlated health indexes
JP7485013B2 (en) Information presentation device, information presentation method, and program
JP7393518B2 (en) Teacher data collection device, teacher data collection method and recording medium

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: 18907614

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 25/11/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18907614

Country of ref document: EP

Kind code of ref document: A1