JP7300973B2 - Healthcare data analysis system and healthcare data analysis method - Google Patents

Description

The present invention relates to a healthcare data analysis system and method for analyzing healthcare information such as medical care and nursing care.

In recent years, while healthy life expectancy and working life expectancy have been extended, the increase in medical and nursing care costs has become a matter of concern due to demographic changes such as the declining birthrate and aging population. Therefore, there is a demand for appropriate measures to prevent the occurrence and aggravation of health, medical care, and nursing care.

In order to decide measures for medical care, it is necessary to examine future predictions and past performance. For example, Patent Literature 1 is known as a technique for accurately predicting a future health condition. In Patent Literature 1, an input vector representing a subject's medical examination record is normalized, and then the input vector is input to a learning algorithm to obtain an output vector that indicates a prediction of the subject's future state.

Further, as a method for examining past performance, Patent Document 2 is known. Patent Literature 2 discloses a technique for obtaining a data expression that enables appropriate clustering of medical data in which different types of data such as health checkup results and interview results are mixed by using latent topic analysis. It is

JP 2018-194904 A JP 2017-027307 A

With the above-described conventional technology, although it is possible to predict the health condition and classify medical information, it is difficult to support planning of measures for medical care, health checkups, or nursing care.

In particular, when predictions and classifications are made from healthcare data for measures implemented in specific regions (municipalities and regions) or occupational areas (groups), the analysis results can be applied to those regions, but When measures implemented in one region are implemented in a different region, the same effects may not always be obtained, and the prediction models and classification methods that are the results of analysis in the original region may not always be applicable. .

In other words, because the environment (climate, climate), demographic composition, and industrial structure differ from region to region, the effects of measures related to health, medical care, prevention of long-term care, etc. are not always the same. For this reason, with the above-described conventional technology, it is difficult to support the creation of nationwide medical measures and measures for other regions from prediction models and classification methods that are the results of analyzing data from some regions. There was a problem.

Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to support the creation of new measures for medical care, health checkups, or nursing care by using healthcare data for measures in regions and the like.

The present invention is a health care data analysis system for analyzing health care data, wherein a computer system having a processor and a memory comprises: a first database managing first health care data of a first group; a knowledge vector generating unit that receives one set of health and medical data, converts the mutual relationship between the first health and medical data into a vector, and outputs the knowledge vector information; A third database that manages a database, second health and medical data of a second group, and policy information about the policy implemented in the second group, and the second health of the third database a first vector transformation for generating vector information by inputting medical data and knowledge vector information of the second database and vector-transforming the second health care data of the second group using the knowledge vector information; a machine learning unit that inputs the converted vector information and the policy information of the third database and outputs model information obtained by modeling the relationship between the vector information and the policy information by machine learning; a fourth database that manages the output model information, wherein the second group is composed of groups subordinate to the first group, and the measure information is stored in the second group; target population of health care measures in the country and the results of the measures.

Therefore, according to the present invention, vector information is calculated from nationwide (first group) health care data, and this vector information is combined with measures and measures implemented in a specific region (second group). Generate model information (prediction model) from vector information of healthcare data. By applying healthcare data from another region (third group) to this model information, when measures taken in a specific region (second group) are implemented in another region (third group) It becomes possible to predict the effect of the above, and it becomes possible to support the creation of new measures for healthcare (medical care, medical examination, or nursing care).

The details of at least one implementation of the subject matter disclosed in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the disclosed subject matter will become apparent from the following disclosure, drawings, and claims.

It is a figure which shows the Example of this invention and shows the outline|summary of a healthcare data analysis system. 1 is a block diagram showing an embodiment of the present invention and showing an example of the configuration of a healthcare data analysis system; FIG. 4 is a flow chart showing an embodiment of the present invention and showing an example of knowledge vector information creation processing performed by the knowledge creation device; 4 is a flow chart showing an embodiment of the present invention and showing an example of model generation processing performed by the model generation device. FIG. 10 is a flow chart showing an embodiment of the present invention and showing an example of simulation processing performed by a simulation device #2; FIG. 4 is a flow chart showing an embodiment of the present invention and showing an example of simulation processing performed by simulation device #1. FIG. 4 shows an embodiment of the present invention and shows an example of health checkup information in a database DB#1; FIG. 4 shows an embodiment of the present invention and shows an example of health checkup information in a database DB#3; FIG. 10 shows an embodiment of the present invention and shows an example of health checkup information in database DB#5; FIG. 4 is a diagram showing an embodiment of the present invention and showing an example of receipt information in a database DB#1; FIG. 10 is a diagram showing an embodiment of the present invention and showing an example of receipt information in a database DB#3; FIG. 10 is a diagram showing an embodiment of the present invention and showing an example of receipt information in a database DB#5; It is a figure which shows the Example of this invention and shows an example of the care information of database DB#1. It is a figure which shows the Example of this invention and shows an example of the care information of database DB#3. It is a figure which shows the Example of this invention and shows an example of the care information of database DB#5. FIG. 10 is a diagram showing an embodiment of the present invention and showing an example of policy information in database DB#3; FIG. 10 shows an embodiment of the present invention and shows an example of policy effect information in database DB#3; FIG. 4 is a diagram showing an embodiment of the present invention and showing an example of matching information generated by the knowledge creation device; FIG. 4 is a diagram showing an embodiment of the present invention and showing an example of vector information generated by the knowledge creation device; FIG. 4 shows an embodiment of the present invention and shows an example of knowledge vector information stored in database DB#1; FIG. 4 shows an embodiment of the present invention and shows an example of model information stored in a database DB#4; FIG. 10 is a diagram showing an example of a model selection screen, showing the embodiment of the present invention; It is a figure which shows the Example of this invention and shows an example of a simulation result screen. It is a block diagram which shows the Example of this invention and shows an example of a structure of a simulation apparatus. FIG. 4 illustrates an embodiment of the present invention and illustrates an example of generation of a vector representation of terms using a Skip-gram model;

An embodiment of the present invention will be described below with reference to the accompanying drawings.

<Overview>
FIG. 1 shows an embodiment of the present invention and is a diagram showing an overview of a healthcare (medical care) data analysis system. The healthcare data of this embodiment is information including at least one of medical information, health checkup information, and nursing care information.

In the healthcare data analysis system, healthcare data collected by country (or higher group) A1 is stored in a database (DB#1) 1 including medical receipt information, health checkup information, and nursing care information. . The health care data collected by country A1 is an anonymous database, and individuals cannot be identified. .

At the level of country A1, as will be described later, knowledge vector information 21 and vector information 23 are generated by vector-converting healthcare data and stored in database (DB#2) 2 . Distributed representation can be used as an example of vector conversion.

On the other hand, the lower region (or lower group) A2 that constitutes the country (or group) A1 is the health care data (medical receipt information, health checkup information, nursing care information) of individuals living or working in the region A2. , information on measures implemented in the area is stored in the database (DB#3) 3 .

In the area A2, the health care data in the database 3 and the policy information are vectorized using the knowledge vector information 21 possessed by the country A1, as will be described later. Then, in the region A2, machine learning is performed on the vectorized information to generate model information 41 for predicting healthcare, and the model information 41 is registered in the database (DB#4) 4 managed by the country A1. The model information 41 is generated for each measure information implemented in the area A2, and can include the result (effect) of the implementation of the measure.

Another area (lower group) A3 that constitutes the country (or group) A1 stores health care data of individuals living or working in the area A3 in the database (DB#5) 5 .

A database 4 managed by country A1 is shared by area A2 and another area A3. Model information 41 stored in database 4 is used in country A1 and other regions A3. In country A1, the model information 41 selected from database 4 can be used to simulate the health care data in database 1 to predict the effects of nationwide implementation of measures implemented in region A2.

Similarly, in another region A3, the model information 41 selected from the database 4 is used to simulate the health care data in the database 5 to predict the effect of implementing the measures implemented in the region A2 in the region A3. be able to.

As described above, in the healthcare data analysis system of this embodiment, the nationwide health care data owned by country A1 is vector-converted to generate knowledge vector information 21 and vector information 23, and the area A2 where the policy is implemented is generated. Model information 41 (prediction model) is generated by machine-learning vectorized information using knowledge vector information 21 for healthcare data and policy information. The vector information 23 is the substance of vector values, and the knowledge vector information 21 is a table for managing the vector information 23. FIG.

In country A1 and another region A3, by using this model information 41 to perform simulations with their respective healthcare data, it is possible to see the effects of implementing various measures implemented in region A2 in country A1 and another region A3. It is possible to predict the effect.

In country A1 and other regions A3, based on the results of the simulation, it is possible to formulate new measures and apply existing measures, thereby efficiently planning and determining healthcare measures. In particular, country A1 and region A3 do not have data on the results of measures implemented in region A2. It is possible to predict the effect of implementing the measures. Furthermore, when creating a prediction model for the region A2, the data of the region A2 is vectorized using the knowledge vector information 21 created from the characteristics of the entire data such as the country A1. As a result, the prediction model created for the region A2 can create a model that reflects not only the data characteristics of the region A2 but also the trends in the overall data of the country A1. Then, when applying the model created in region A2 to country A1 or region A3, the data of country A1 or region A3 are vectorized with the same knowledge vector information 21 and applied to the prediction model, so that country A1 There is an effect that it is possible to accurately predict the case where it is applied to the area A3. In this way, the healthcare data analysis system of this embodiment can support the planning and determination of healthcare measures.

In the above example, the top group is country A1, and the bottom groups (A2, A3) are regions. However, the top group (A1) is the entire company. and the groups (A2, A3) can be business departments.

<System configuration>
FIG. 2 is a block diagram showing an example configuration of a healthcare data analysis system. The healthcare data analysis system of this embodiment includes data center DC#1 managed by country (group) A1, data center DC#2 managed by region (group) A2, and data managed by region (group) A3. An example in which the center DC#3 is configured by a network (not shown) is shown.

A data center DC#1 managed by country (group) A1 includes a database 1 that stores anonymous healthcare data, a knowledge creation device 9 that creates knowledge vector information 21 and vector information 23, and A database 4 that stores model information 41, and a simulation device (#1) 4 that selects the model information 41 from the database 4 and performs a simulation using the healthcare data of the database 1 and the knowledge vector information 21 and vector information 23. .

Data center DC#2 managed by area (group) A2 includes database 3 storing health care data including personal information residing in area A2 and policy information, and knowledge vector information 21 and vector information 23 of data center DC#1. a model generation device 7 that vectorizes healthcare data and policy information of the area A2 using , generates a prediction model, and stores it as model information 41 in the database 4 of the data center DC#1.

Data center DC#3 managed by region (group) A3 stores database 5 storing healthcare data including personal information residing in region A3, and model information 41 and knowledge selected from database 4 of data center DC#1. and a simulation device 8 that uses the vector information 21 and the vector information 23 to perform simulations with healthcare data in the database 5 .

In the database 1 of the data center DC#1, health checkup information 11 that anonymously stores information such as health checkups, medical bill information 12 that anonymously stores medical bills from medical institutions, nursing care facilities, etc. are used. Nursing care information 13 that anonymously stores information at the time of care is stored.

The knowledge creation device 9 is a computer including a CPU 91, a memory 92, and a communication device (not shown). A knowledge vector generator 93 is loaded as a program into the memory 92 and executed by the CPU 91 .

The knowledge vector creation unit 93 performs vector conversion using distributed representation using any of the health checkup information 11, the receipt information 12, and the nursing care information 13 in the database 1, generates the vector information 23 and the knowledge vector information 21, and stores the information in the database 2. store to In this embodiment, the number of dimensions for vectorization of the vector information 23 is m.

Although the health checkup information 11, the health insurance claim information 12, and the nursing care information 13 in the database 1 are anonymized, the same individual ID (identifier) is given to the same person. and the individual IDs of the nursing care information 13, and after combining each information (table), vector conversion can be performed.

The simulation device (#1) 6 is a computer including a CPU 61, a memory 62, an input device 67, an output device 68, a communication device 69, and a storage device 66, as shown in FIG.

The input device 67 is composed of a keyboard, a mouse, a touch panel, or the like. The output device 68 consists of a display. The communication device 69 communicates with other computers via a network (not shown). The storage device 66 is composed of a non-volatile storage medium and stores data and programs.

The output unit 63 , the vector conversion unit 64 , and the estimation calculation unit 65 are loaded as programs into the memory 62 and executed by the CPU 61 .

The simulation device (#1) 6 selects from a plurality of model information 41 stored in the database 4 by the user, and vectorizes the healthcare data and the knowledge vector information 21 in the database 1 by the vector conversion unit 64. , the model information 41 selected by the estimation calculation unit 65 and the vector information 23 are used for simulation.

The vector conversion unit 64 vectorizes at least one of the medical insurance claim information 12, health checkup information 11, and nursing care information 13 in the database 1 selected by the user. The vector conversion unit 64 can match the table selected by the user by personal ID and then vectorize the table.

The estimation calculation unit 65 receives the vectorized data and the vector information 23 of the knowledge vector information 21 , performs simulation with the selected model information 41 , and outputs the result to the output device 68 .

The CPU 61 operates as a functional unit that provides a predetermined function by executing processing according to the program of each functional unit. For example, the CPU 61 functions as the vector conversion unit 64 by executing processing according to a vector conversion program. The same is true for other programs. Furthermore, the CPU 61 also operates as a functional unit that provides functions of a plurality of processes executed by each program. Computers and computer systems are devices and systems that include these functional units.

Next, a data center DC#2 managed by an area (group) A2 that constitutes a country (group) A1 includes a database 3 and a model generation device 7. FIG. In the database 3, as health care data of individuals who live or work in the area A2, medical insurance claim information 32, health checkup information 31, nursing care information 33, policy information 34 implemented in the area, and policy effect information 3410. , are stored. The database 3 of the area A2 stores healthcare data including personal information, unlike the database 3 of the country A1.

Health checkup information 31 stores information such as health checkups, medical insurance claim information 32 stores medical insurance claims, and nursing care information 33 stores information when nursing care facilities and the like are used.

The policy information 34 stores the content of the policy implemented in the area A2. As will be described later, the policy information 34 stores the subject of the health care policy and the results of the policy (the effect of the policy).

The model generation device 7 is a computer including a CPU 71, a memory 72, and a communication device (not shown). A vector conversion unit 73 and a machine learning unit 74 are loaded as programs into the memory 72 and executed by the CPU 71 .

The vector conversion unit 73 vectorizes the health care data (medical checkup information 31, claim information 32, nursing care information 33) and the policy information 34 in the database 3 using distributed representation. A prediction model is generated by learning and stored as model information 41 in the database 4 of the data center DC#1.

Next, a data center DC#3 managed by an area (group) A3 that constitutes a country (group) A1 includes a database 5 and a simulation device (#2) 8. FIG. The database 5 stores medical insurance claim information 52, health checkup information 51, and nursing care information 53 as healthcare data for individuals who live or work in the area A3. The database 5 of region A3 is health care data including personal information, unlike country A1.

Health checkup information 51 stores information such as health checkups, medical bill information 52 stores medical bills received at medical institutions, and nursing care information 53 stores information when nursing care facilities and the like are used.

The simulation device (#2) 8 is a computer including a CPU 81, a memory 82, and a communication device (not shown). An output unit 83 , a vector conversion unit 84 , and an estimation calculation unit 85 are loaded as programs into the memory 82 and executed by the CPU 81 .

In the simulation device (#2) 8, the user selects from a plurality of model information 41 accumulated in the database 4 of the data center DC#1, and the healthcare data in the database 5 is vectorized by the vector conversion unit 84. Therefore, a simulation is performed using the model information 41 selected by the estimation calculation unit 85 and the vector information 23 of the knowledge vector information 21 .

The vector conversion unit 84 vectorizes at least one of the medical insurance claim information 52, health checkup information 51, and nursing care information 53 in the database 1 selected by the user. The vector conversion unit 84 can match the table selected by the user by personal ID and then vectorize the table.

The estimation calculation unit 85 receives the vectorized data and the vector information 23 of the knowledge vector information 21, performs a simulation with the selected model information 41, and outputs the result to an output device (not shown). .

Note that FIG. 2 shows an example in which two regions constitute country A1, but the present invention is not limited to this. Also, an example has been shown in which the model generation device 7 is arranged in the area A2 and the simulation device 8 is arranged in the area A3, but the present invention is not limited to this. For example, the model generation device 7 and the simulation device 8 are placed in each of the regions A2 and A3 to generate the model information 41 based on the policy and healthcare data of the region and register it in the database 4 of the data center DC#1. , the simulation can be performed with the model information 41 of other regions.

Also, the knowledge creation device 9, the model generation device 7, and the simulation device (#2) 8 have an input device and an output device like the simulation device (#1) 6 shown in FIG. is received by the input device, and the processing result is displayed on the output device.

<Data>
Next, data used in each data center DC#1 to DC#3 will be described. FIG. 7 is a diagram showing an example of health checkup information 11 in the database (DB#1) 1. As shown in FIG. Health checkup information 11 is a table managed by country A1.

The health checkup information 11 includes a health checkup ID 110, an area ID 111, an individual ID 112, an examination date 113, an age at the time of reception 114, an abdominal circumference 115, a diastolic blood pressure 116, a systolic blood pressure 117, a blood sugar level 118, and a blood sugar level 118. Sexual fat 119 is included in one record.

The health checkup ID 110 stores an identifier that indicates the institution that performed the health checkup. The area ID 111 stores an identifier indicating the area where the health checkup was performed. The individual ID 112 is given an identifier of the examinee. Waist circumference 115 to neutral fat 119 store measured values.

The database 1 managed by country A1 is an anonymous database and does not contain information such as names and addresses that identify individuals, and the personal ID 112 functions as information that identifies whether or not it is the same person.

Next, FIG. 8 is a diagram showing an example of medical checkup information 31 in the database (DB#3) 3. As shown in FIG. The health checkup information 31 is a table managed by the region A2 and includes personal information that identifies an individual.

The health checkup information 31 includes a health checkup ID 310, an area ID 311, an individual ID 312, an examination date 313, an age at reception 314, an abdominal circumference 315, a diastolic blood pressure 316, a systolic blood pressure 317, a blood sugar level 318, and a blood sugar level 318. Sexual fat 319 is included in one record.

The health checkup ID 310 stores an identifier that indicates the institution that performed the health checkup. The area ID 311 stores an identifier indicating the area where the health checkup was performed. The individual ID 312 is given an identifier of the examinee.

The database 3 managed by the area A2 is a database containing personal information such as names and addresses, and the personal ID 112 is associated with the personal information. Measurement values are stored in waist circumference 315 to neutral fat 319 .

Next, FIG. 9 is a diagram showing an example of health checkup information 51 in the database (DB#5) 5. As shown in FIG. The health checkup information 51 is a table managed by another area A3 and includes personal information that identifies an individual.

The health checkup information 51 includes a health checkup ID 510, an area ID 511, an individual ID 512, an examination date 513, an age at the time of reception 514, an abdominal circumference 515, a minimum blood pressure 516, a maximum blood pressure 517, a blood sugar level 518, and a blood sugar level 518. Sexual fat 519 is included in one record.

The health checkup ID 510 stores an identifier that indicates the institution that performed the health checkup. The area ID 511 stores an identifier indicating the area where the health checkup was performed. The individual ID 512 is given an identifier of the examinee.

The database 3 managed by the area A2 is a database containing personal information such as names and addresses, and the personal ID 112 is associated with the personal information. Measurement values are stored in waist circumference 515 to neutral fat 519 .

FIG. 10 is a diagram showing an example of the receipt information 12 of the database (DB#1) 1. As shown in FIG. Receipt information 12 is a table managed by country A1.

The medical insurance claim information 12 includes a medical insurance claim ID 120, an area ID 121, an individual ID 122, a medical treatment date 123, a medical institution number 124, an injury or disease name 125, a medical treatment 126, a medicine 127, and a billing point 128. in one record.

The receipt ID 120 stores the identifier of the medical bill. The region ID 111 stores an identifier indicating the region where medical care was performed. The individual ID 112 is given an identifier of the examinee.

The medical institution number 124 stores the identifier of the medical institution that provided medical care. The disease name 125 stores the name or code of the disease. The medical practice 126 stores the code of the medical treatment applied to the patient. The medicine 127 stores prescribed medicines. The claimed score 128 stores health scores.

The database 1 managed by the country A1 is an anonymous database and does not contain information such as names and addresses that identify individuals.

Next, FIG. 11 is a diagram showing an example of the receipt information 32 of the database (DB#3) 3. As shown in FIG. Receipt information 32 is a table managed by region A2 and includes personal information that identifies an individual.

The medical insurance claim information 32 includes a medical insurance claim ID 320, an area ID 321, an individual ID 322, a medical treatment date 323, a medical institution number 324, an injury or disease name 325, a medical treatment 326, a medicine 327, and a billing point 328. in one record.

The receipt ID 320 stores the identifier of the medical bill. The region ID 111 stores an identifier indicating the region where medical care was performed. The individual ID 132 is given an identifier of the examinee.

The medical institution number 324 stores the identifier of the medical institution that provided medical treatment. The disease name 325 stores the name or code of the disease. The medical practice 326 stores the code of the medical care content applied to the patient. The medicine 327 stores prescribed medicines. The claimed score 328 stores health scores.

The database 3 managed by the region A2 is a database containing personal information such as names and addresses, and the personal ID 322 is associated with the personal information.

Next, FIG. 12 is a diagram showing an example of the receipt information 52 of the database (DB#5) 5. As shown in FIG. Receipt information 52 is a table managed by other area A3 and includes personal information that identifies an individual.

The medical insurance claim information 52 includes a medical insurance claim ID 520, an area ID 521, an individual ID 522, a medical treatment date 523, a medical institution number 524, an injury or disease name 525, a medical treatment 526, a medicine 527, and a billing point 528. in one record.

The receipt ID 520 stores the identifier of the medical bill. The region ID 111 stores an identifier indicating the region where medical care was performed. The individual ID 152 is given an identifier of the examinee.

The medical institution number 524 stores the identifier of the medical institution that provided medical treatment. The disease name 525 stores the name or code of the disease. The medical practice 526 stores the code of the medical treatment applied to the patient. The medicine 527 stores prescribed medicines. The claimed score 528 stores health scores.

The database 5 managed by the other area A3 is a database containing personal information such as name and address, and the personal ID 522 is associated with the personal information.

FIG. 13 is a diagram showing an example of nursing care information 13 in the database (DB#1) 1. As shown in FIG. Nursing care information 13 is a table managed by country A1.

Nursing care information 13 includes nursing care information ID 130, region ID 131, personal ID 132, year and month 133, level of care required 134, office number 135, service code 136, and billing amount 137 in one record. include.

Nursing care information ID 130 stores an identifier of a nursing care request. The area ID 131 stores an identifier indicating the area where the nursing care service was provided. The individual ID 132 is assigned a caregiver identifier. The year and month 133 stores the year and month when the care service was received. The degree of need for care 134 stores the degree of need for care determined by certification of need for care.

The office number 135 stores the identifier of the office that provided the nursing care service. The service code 136 stores the code of the implemented care service. The amount billed 137 stores the amount billed for the care service.

The database 1 managed by country A1 is an anonymous database and does not contain information such as names and addresses that identify individuals.

FIG. 14 is a diagram showing an example of nursing care information 33 in database (DB#3) 3. As shown in FIG. Nursing care information 33 is a table managed by area A2.

Nursing care information 33 includes nursing care information ID 330, region ID 331, personal ID 332, year/month 333, level of care required 334, office number 335, service code 336, and billing amount 337 in one record. include.

Nursing care information ID 330 stores an identifier of a nursing care request. The area ID 331 stores an identifier indicating the area where the nursing care service is provided. The personal ID 332 is assigned a caregiver identifier. The year and month 333 stores the year and month when the care service was received. The degree of need for care 334 stores the degree of need for care determined by certification of need for care.

The office number 335 stores the identifier of the office that provided the nursing care service. The service code 336 stores the code of the implemented care service. The amount billed 337 stores the amount billed for the care service.

The database 3 managed by the area A2 is a database containing personal information such as names and addresses, and the personal ID 332 is associated with the personal information.

FIG. 15 is a diagram showing an example of nursing care information 53 in database (DB#5) 5. As shown in FIG. Nursing care information 53 is a table managed by another area A3.

Nursing care information 53 includes nursing care information ID 530, area ID 531, personal ID 532, year/month 533, nursing care level 534, office number 535, service code 536, and billing amount 537 in one record. include.

The care information ID 530 stores an identifier of care information. The area ID 531 stores an identifier indicating the area where the care service was provided. The individual ID 532 is assigned a caregiver identifier. The year and month 533 stores the year and month when the care service was received. The degree of need for care 534 stores the degree of need for care determined by certification of need for care.

The office number 535 stores the identifier of the office that provided the care service. The service code 536 stores the code of the implemented care service. The amount billed 537 stores the amount billed for the care service.

The database 5 managed by the other area A3 is a database containing personal information such as names and addresses, and the personal ID 532 is associated with the personal information.

FIG. 16A is a diagram showing an example of policy information 34 in database (DB#3) 3. FIG. The policy information 34 is a table managed by the area A2.

The policy information 34 includes an area ID 340, a policy ID 341, a policy name 342, a target person ID 343, and a goal achievement 344 in one record. The region ID 340 stores the identifier of the region where the policy was implemented. The policy ID 341 stores an identifier for identifying the policy.

The policy name 342 stores the name of the policy. The target person ID 343 stores an individual ID that participated in the policy. The goal achievement 344 stores a value indicating whether or not the target person has achieved the goal of the policy, and stores "1" if the target has been achieved and "0" if the target has not been achieved. be.

The target person ID 343 of the policy information 34 corresponds to the personal ID of the health checkup information 31 , the receipt information 32 or the care information 33 of the database 3 . The policy information 34 can manage whether or not the individual participating in the policy has reached the target set for each policy.

FIG. 16B is a diagram showing an example of policy effect information 3410 in database (DB#3) 3. As shown in FIG. The policy effect information 3410 is a table managed by the region A2.

The measure effect information 3410 includes a measure ID 3411, a measure name 3412, an implementation year 3413, an effect index 3414, and a measured effect 3415 in one record. The measure ID 3411 stores the identifier of the measure and corresponds to the measure ID 341 of the measure information 34 .

The policy name 3412 stores the name of the policy and corresponds to the policy name 342 of the policy information 34 . The implementation year 3413 stores the year in which the measure was implemented. The effect index 3414 stores an index for measuring the effect of the policy. The measured effect 3415 stores the index value set in the effect index 3414 . The policy effect information 3410 manages the measured effect 3415 for each policy.

FIG. 17 is a diagram showing an example of the matching information 22 generated by the knowledge creation device 9 of the data center DC#1. The knowledge creation device 9 can create the vector information 23 of the knowledge vector information 21 for the health checkup information 11, the claim information 12, and the care information 13 independently. may be combined before generating the knowledge vector information 21 and the vector information 23 .

The example of FIG. 17 shows an example in which health checkup information 11, medical insurance claim information 12, and nursing care information 13 are combined by matching personal IDs 112, 122, and 132. FIG. Values of individual IDs 112 , 122 , and 132 are stored in case 221 .

The receipt information 222 stores the value of the record of the receipt information 12 corresponding to the individual ID (122) of the case 221. FIG. In the care information 223, the value of the record of the care information 13 corresponding to the individual ID (132) of the case 221 is stored. The health checkup information 224 stores the value of the record of the health checkup information 11 corresponding to the individual ID (112) of the case 221. FIG.

The knowledge creation device 9 generates matching information 22 by matching and combining medical checkup information 11, medical insurance claim information 12, and nursing care information 13 by personal ID, and then performs vector conversion. Note that, in this embodiment, the number n of columns of the receipt information 222, the nursing care information 223, and the health checkup information 224 of the matching information 22 is set to the number of dimensions n before vector conversion.

When generating the vector information 23 , the knowledge creation device 9 designates the data source read from the database 1 , year (or period), region, etc., and sets these designations in the knowledge vector information 21 .

FIG. 18 is a diagram showing an example of the vector information 23 generated by the knowledge creation device 9. As shown in FIG. The vector information 23 is information obtained by converting the matching information 22 into a vector value using distributed representation by the knowledge vector generation unit 93 .

The vector information 23 includes a knowledge ID 231, an item 232, and a vector value 233 in one record. The knowledge ID 231 stores identifiers indicating groups managed by the knowledge vector information 21 . The item 232 stores the field names of the matching information 22 (item names of the receipt information 222, the nursing care information 223, and the health checkup information 224).

The illustrated example shows an example in which the n-dimensional matching information 22 shown in FIG. 17 is vectorized by m-dimensional distributed representation.

FIG. 19 is a diagram showing an example of knowledge vector information 21 generated by the knowledge creation device 9. As shown in FIG. The knowledge vector information 21 includes knowledge ID 211, data source 212, year 213, and area 214 in one record.

The knowledge ID 211 is an identifier indicating the group of the vector information 23 given by the knowledge vector creation unit 93, and corresponds to the knowledge ID 231 in FIG. The data source 212 stores the information of the database 1 that is input when the vector value of the knowledge ID 211 is generated. The fiscal year 213 stores the fiscal year in which the data was selected from the data source of the database 1 . The area 214 stores the area for which data is selected from the data source of the database 1 .

The knowledge vector information 21 manages the data source, year and region of the vector information 23 by using knowledge IDs 231 for groups of the vector information 23 . When using the vector information 23 from the simulation devices 6 and 8, the knowledge vector information 21 is referred to, the desired knowledge ID 211 is selected from the data source 212, the year 213, the area 214, etc., and the selected knowledge ID 211 The record of the knowledge ID 231 of the vector information 23 that matches the value of is obtained.

In this embodiment, an example is shown in which the knowledge vector information 21 and the vector information 23 are separated, but the data source 212, the year 213, and the region 214 may be added to the vector information 23 to form one table.

FIG. 20 is a diagram showing an example of model information 41 generated by the model generation device 7. As shown in FIG. The model information 41 includes a model ID 411, a model 412, a knowledge ID 413, a policy ID 414, and a track record 415 in one record.

The model ID 411 stores an identifier set by the model generation device 7 . The model 412 stores the model generated by the model generation device 7 . The knowledge ID 413 stores the value of the knowledge ID 211 of the knowledge vector information 21 acquired from the database 2 at the time of model generation.

The measure ID 414 stores the identifier of the measure implemented in the area A2. The policy ID 414 corresponds to the policy ID 341 of the policy information 34 in FIG. 16A. The results 415 store the results of implementing measures. The result 415 corresponds to the goal achievement 344 in FIG. 16A. “Target unachieved” is set in the achievement 415 .

By referring to the model information 41, the simulation devices 6 and 8 can determine which measures have achieved their goals and which have not.

<Processing>
FIG. 3 is a flow chart showing an example of processing for creating the knowledge vector information 21 performed by the knowledge creation device 9 of the data center DC#1. This processing is started in response to an instruction from the user of the knowledge creation device 9. FIG.

First, the knowledge creation device 9 acquires data specified by a user from healthcare data (health and medical information) in the database 1 (S1). The knowledge creation device 9 receives a table to be read from the database 1, a year (or period), a target area, etc., and acquires data (health care data) of specified contents from the database 2. FIG.

When a plurality of data sources are specified, the knowledge creation device 9 matches individual IDs (case IDs 221) to generate matching information 22 in which a plurality of tables are combined, as shown in FIG.

Next, the knowledge vector creation unit 93 of the knowledge creation device 9 vectorizes the mutual relationship between the health and medical data from the data source (matching information 22) acquired from the database 2 by distributed representation, and generates the vector information 23 as follows: Generate (S2). As shown in FIG. 18 , the knowledge creation device 9 calculates m-dimensional vector values for n-dimensional items 232 and assigns knowledge IDs 232 to each item 232 .

The knowledge creation device 9 creates the knowledge vector information 21 from the input information such as the knowledge ID 231 of the created vector information 23, data source, year and region. Then, the knowledge creation device 9 stores the generated knowledge vector information 21 and vector information 23 in the database 2 (S3).

Skip-gram model and the like are known as an example of a method of vectorizing mutual relationships between health and medical data using distributed representation. In this embodiment, too, the Skip-gram model can be used to generate the vector representation of terms.

FIG. 24 is a schematic diagram showing an example of generating a vector representation of terms using the Skip-gram model.

FIG. 24 shows a schematic diagram of the model used in this example. The knowledge vector generator 93 sets the initial values of the parameters of this model. First, the model used in this embodiment will be described below with reference to FIG. In this embodiment, a two-layer model shown in FIG. 24 is used.

A one-hot vector 401 indicated by x on the left side of FIG. 24 is a vector in which only one element indicating a term is 1 and all other elements are 0. FIG. Here x is a column vector. The number of dimensions is equal to the number of terms for which the term representation vector is created. That is, let z1, z2, . The vector x=(0, . . . , 0, 1, 0, . . . , 0)^T. where x^T represents the transpose of the vector x.

For the sake of simplicity, if there are three target terms, drug A, drug B, and drug C, the vector x is three-dimensional. If the corresponding elements of drug A, drug B, and drug C are the first element, the second element, and the third element, drug A is x = (1, 0, 0)^T, and drug B is x = ( 0, 1, 0)^T, drug C is expressed as x = (0, 0, 1)^T.

In the above description, three examples of terms are shown, but in reality, the number of names of injuries and diseases, names of medicines, and names of medical procedures is enormous, so the number of dimensions of the one-hot vector 401 is also large.

Now, one record is selected from the data source (matching information 22) acquired from the database 2. FIG. Let w1, w2, . Also, let x1, x2, . At this time, FIG. 24 shows one-hot vectors x1, . . . , xi-1, . It is a model that predicts xi+1, . . . , xT. The matrices W and W' are the parameters of this model, and these parameters are updated to improve the prediction accuracy based on the receipt data.

In FIG. 24, first, the n-dimensional one-hot vector xi of the term wi is transformed by the m×n matrix W into an m-dimensional vector Wxi. m is the number of dimensions of the term expression vector and is determined in advance. It is preferable to set the size to about several hundred dimensions. m is generally a small value compared to n, which is equal to the total number of terms, and Wxi can be regarded as an m-dimensional vector encoding of xi. Using this, recover the n-dimensional one-hot vectors x1,..., xi-1, xi+1,..., xT of the terms w1,..., wi-1, wi+1,..., wT co-occurring with the term represented by xk. think about For this purpose, the n×m matrix W' is converted into an n-dimensional vector W'Wx.

Here, the inner product is calculated as the similarity between the one-hot vector xj of wj and W'Wxi. Assume that xi is 1 in the Ii-th element and 0 in the other elements. Also, using an m-dimensional column vector vk, let W=(v1, v2, . . . , vn). Also, using the m-dimensional column vector v'k, let W'=(v'1, v'2, . . . , v'n)^T. At this time, Wxi=vIi. It is also assumed that the Ij-th element of xj is 1 and the other elements are 0. At this time, the inner product of xj and W'Wxj is equal to the inner product v'Ij·vIi of v'Ij and vIi. Thus, in order to calculate the inner product of xj and W'Wxi, the inner product v'Ij·vIi of m-dimensional vectors should be calculated.

Now, the softmax function is used to define as follows. where Σ_p adds from 1 to n for p. Since this calculation needs to be performed for all terms, the amount of calculation is large. In practice, Hierarchical Softmax or NEG (Negative Sampling) should be used.
p(wj|wi)=exp(v'Ij·vIi)/Σ_p exp(v'p·vIi)

The model uses the matching information 22 to update W and W' to maximize the function In the following, Σ_i is the addition of 1 to T for i, and Σ_j≠i is the addition of 1 to T except i for j.
(1/T) Σ_i Σ_j≠i log {p(wj|wi)}

For each match 22, we iteratively update W and W' by gradient descent to maximize the above log-probability value. At this time, when the one-hot vector of the term w is a vector x in which the i-th value is 1 and the other elements are 0, Wx=vi is the term expression vector of the term w.

Here, the first layer is called the input layer, the second layer is called the output layer, W is the input layer matrix, W' is the output layer matrix, x is the input vector, and Wx is the hidden vector.

Through the above processing, vector information 23 having m-dimensional vector values and knowledge vector information 21 as catalog information of the vector information 23 are generated from the specified healthcare data and stored in the database 2 .

FIG. 4 is a flowchart showing an example of model generation processing performed by the model generation device 7 of the data center DC#2. This processing is started in response to an instruction from the user of the model generating device 7. FIG.

The model generation device 7 generates knowledge vector information 21 selected by the user from the database 2 of the data center DC#1, and healthcare data information (health checkup information 31, Receipt information 32 or care information 33) and policy information 34 are received (S11).

The model generation device 7 acquires from the vector information 23 the data of the knowledge ID 231 corresponding to the knowledge ID 211 of the knowledge vector information 21 of the data center DC#1. The model generation device 7 also reads from the database 3 the table selected by the user and the healthcare data corresponding to the year and region. When a plurality of healthcare data are specified, the model generation device 7 generates matching information 22 in which the tables are combined by performing matching using individual IDs (case IDs 221) as shown in FIG. . The model generation device 7 also acquires the policy ID 414 of the policy information 34 selected by the user.

Next, the vector conversion unit 73 of the model generation device 7 uses the obtained vector information 23 to vectorize the obtained healthcare data (medical checkup information 31, receipt information 32, nursing care information 33, or matching information 22). (S12).

For example, the vector conversion unit 73 gives an m-dimensional vector value 233 to an item of healthcare data that matches an item of the vector information 23, and calculates the vector value by distributed representation. Note that the vectorization of the health care data (health checkup information 31, receipt information 32, nursing care information 33) in the database 3 is not limited to this, and vectorization is performed using the vector information 23 (knowledge vector information 21). A well-known or publicly known technique may be applied.

Next, the machine learning unit 74 of the model generating device 7 receives the vectorized healthcare data and generates a model of the relationship between the healthcare data and the measure information 34 (S13). The machine learning unit 74 performs learning and generates a model using, for example, the ratio of target achievement 344 of the policy information 34 (ratio of “1”) and the measured effect 3415 of the policy effect information 3410 as objective variables.

The model generation device 7 stores the generated model in the database 4 of the data center DC#1 as model information 41 (S14), and the model information 41 is generated by the simulation devices 6 and 8 of the data centers DC#1 and DC#2. share it.

By the above process, the model generation device 7 uses the vector information 23 generated in the data center DC#1 to vectorize the healthcare data stored in the database 2 of the data center DC#2, and the machine learning unit 74 Model information 41 can be generated from healthcare data vectorized with designated policy information 34 as an objective variable.

Also, by applying the nationwide vector information 23 to the vectorization of the healthcare data of the area A2, the accuracy of the generated model information 41 (prediction model) can be improved.

FIG. 5 is a flowchart showing an example of simulation processing performed by the simulation device 8 of the data center DC#3. This processing is started in response to an instruction from the user of the simulation device 8. FIG.

The simulation device 8 generates knowledge vector information 21 and model information 41 selected by the user from the database 2 of the data center DC#1, and healthcare data (health checkup information) selected by the user from the database 5 of the data center DC#3. 51, receipt information 52 or care information 53) is received (S21).

The simulation device 8 acquires the vector information 23 of the knowledge ID 231 corresponding to the knowledge ID 211 of the knowledge vector information 21 of the data center DC#1. Also, the simulation device 8 reads from the database 3 the health care data corresponding to the table selected by the user and the year and region.

When a plurality of data sources of the database 5 are specified, the simulation device 8 matches individual IDs (case IDs 221) as shown in FIG. do. Also, the simulation device 8 acquires the selected model information 41 from the database 4 of the data center DC#1. The model information 41 is obtained by using the model information 41 selected by the user of the simulation device 8 from the model selection screen 550 displayed on the output device (not shown).

FIG. 21 is a diagram showing an example of a model selection screen 550 displayed by the simulation device 8 on an output device (not shown). The model selection screen 550 includes an area 551 for entering the code of the area where the model information 41 was generated, a year 552 for entering the period of the database where the model information 41 was generated, a model selection area 560, and a simulation execution button 553. .

The model selection area 560 has a checkbox 561 , a measure ID 562 , a measure name 563 , an implementation year 564 , an area number 565 , an effect index 566 and an actual measurement effect 567 . The user of the simulation device 8 can select the model information 41 by checking the check box 561 of the desired policy ID 562 model.

Note that the measure ID 562 corresponds to the measure ID 414 of the model information 41, the measure name 563 corresponds to the measure name 342 of the measure information 34, the implementation year 564 corresponds to the implementation year 3413 of the measure effect information 3410, and the region The number 565 corresponds to the area ID 340 of the policy information 34 , the effect index 566 corresponds to the effect index 3414 of the policy effect information 3410 , and the measured effect 567 corresponds to the measured effect 3415 of the measure effect information 3410 .

Next, the vector conversion unit 84 of the simulation device 8 vectorizes the obtained healthcare data (or matching information 22) using the obtained vector information 23 (S22). This process is the same as the process of the vector conversion section 73 in step S12.

Next, the estimation calculation unit 85 of the simulation device 8 inputs the vectorized healthcare data (or the matching information 22) and performs a simulation using the acquired model information 41 (S23).

The output unit 83 of the simulation device 8 displays the simulation result of the estimation calculation unit 85 on an output device (not shown). By this processing, the simulation device 8 of the region A3 can estimate the effect of implementing the policy of the region A2 corresponding to the policy ID 414 of the model information 41 with the health care data of the database 5 .

FIG. 22 is a diagram showing an example of a screen of a simulation result 610 displayed on an output device (not shown) by the simulation device 8. As shown in FIG. The simulation result 610 includes an area 611 displaying the code of the area, a year 612 indicating the period of the data source, a graph area 620 displaying the target achievement rate for each measure, and a graph area 630 displaying the restraint amount for each measure. and an area 640 for selecting simulation results to be output to the graph area. Goal achievement refers to the improvement that the policy is aiming for, such as the number of people who achieved the target weight in a weight loss program, the number of people who successfully quit smoking in a smoking cessation program, or the number of people who were able to improve their activity level through rehabilitation or prevention of nursing care. It is the number of people and the percentage of achievement of the target of the indicator.

Area 640 includes check box 641 , measure ID 642 , measure name 643 , applicable number of people 644 , target achievement 645 , effect index 646 , and predicted value 647 . The user of the simulation apparatus 8 can display the graph of the simulation result in the graph areas 620 and 630 by checking the simulation result check box 641 of the desired policy ID 642 .

Note that the policy ID 642 corresponds to the policy ID 414 of the model information 41, the policy name 643 corresponds to the policy name 342 of the policy information 34, and the applicable number of people 644 corresponds to the number of people in the healthcare data of the database 5 used for the simulation. Correspondingly, the goal achievement (prediction) 645 corresponds to the number of people who can achieve the goal among the applicable number of people 644, the effect index 646 corresponds to the effect index 3414 of the measure effect information 3410, and the predicted value 647 is the effect index. It corresponds to a value of 646 (money amount of simulation results). In this way, as shown in Figures 21 and 22, it is possible to see how effective a weight loss program implemented in a certain region would be if it were applied in one's own region (or country), such as achieving weight loss targets and curbing medical expenses. You can check if you can get it.

FIG. 6 is a flowchart showing an example of simulation processing performed by the simulation device 6 of the data center DC#1. This processing is started in response to an instruction from the user of the simulation device 6. FIG.

The simulation device 6 generates knowledge vector information 21 and model information 41 selected by the user from the database 2 of the data center DC#1, and healthcare data (health checkup information) selected by the user from the database 1 of the data center DC#1. 11, receipt information 12 or nursing care information 13) is received (S31).

The simulation device 6 acquires the vector information 23 of the knowledge ID 231 corresponding to the knowledge ID 211 of the knowledge vector information 21 of the data center DC#1. Also, the simulation device 6 reads from the database 1 the health care data corresponding to the table selected by the user and the year and region.

Incidentally, when a plurality of data sources of the database 1 are specified, the simulation device 6 matches individual IDs (case IDs 221) as shown in FIG. do. Also, the simulation device 6 selects the selected model information 41 from the database 4 of the data center DC#1 in the same manner as in FIG.

Next, the vector conversion unit 64 of the simulation device 6 vectorizes the obtained healthcare data (or matching information 22) using the obtained vector information 23 (S32). This process is the same as the process of the vector conversion section 73 in step S12.

Next, the estimation calculation unit 65 of the simulation device 6 inputs the vectorized healthcare data (or matching information 22) and performs a simulation using the acquired model information 41 (S33).

The output unit 63 of the simulation device 6 displays the simulation result of the estimation calculation unit 65 on the output device (not shown), as in FIG. 22 above. By this processing, the simulation device 6 of the country A1 can estimate the effect of implementing the policy for the area A2 corresponding to the policy ID 414 of the model information 41 using the healthcare data of the database 1 .

As described above, the healthcare data analysis system of this embodiment calculates the vector information 23 and the knowledge vector information 21 from nationwide healthcare data, and combines this vector information 23 with the policy information 34 of a specific area A2. Model information (prediction model) 41 is generated from healthcare data. By applying the healthcare data of another region A3 (or country A1) to this model information 41, predicting the effect when the measures of a specific region A2 are implemented in another region A3 (or country A1) becomes possible. This makes it possible to support creation of new measures for health care (medical care, physical examination, or nursing care). In particular, country A1 and region A3 do not have data on the results of measures implemented in region A2. It is possible to predict the effect of implementing the measures. Furthermore, when creating a prediction model for the region A2, the data of the region A2 is vectorized using the knowledge vector information 21 created from the characteristics of the entire data such as the country A1. As a result, the prediction model created for the region A2 can create a model that reflects not only the data characteristics of the region A2 but also the trends in the overall data of the country A1. Then, when applying the model created in region A2 to country A1 or region A3, the data of country A1 or region A3 are vectorized with the same knowledge vector information 21 and applied to the prediction model, so that country A1 There is an effect that it is possible to accurately predict the case where it is applied to the area A3.

In the above embodiment, the data center DC#1 in the country A1 is the highest region (or group), and the data center DC#2 in the region A2 and the data center DC#3 in the region A3 are the lower groups. Although an example with a data center is shown, it is not so limited.

For example, the knowledge creation device 9, the simulation device 6, the model generation device 7, the simulation device 8, and each database may be arranged in one data center. Alternatively, the functions of the knowledge generation device 9, the simulation device 6, the model generation device 7, and the simulation device 8 are realized by one computer, and the knowledge generation unit, the first simulation unit, the model generation unit, and the second simulation unit are realized. good too.

<Conclusion>
As described above, the healthcare data analysis system of the above embodiment can be configured as follows.

(1) A computer system having a processor and a memory is a health care data analysis system that analyzes health care data, and includes first health care data (medical checkup information 11, receipt information) of a first group (country A1) 12, a first database (DB # 1) that manages nursing care information 13) and the first health care data are input, and the mutual relationship between the first health care data is converted into a vector A knowledge vector generator (93) that outputs knowledge vector information (21), a second database (DB#2) that stores the knowledge vector information (21), and a second A third database (DB #3), the second health and medical data of the third database (DB#3), and the knowledge vector information (21) of the second database (DB#2), and the second group A first vector conversion unit (73) for generating vector information by vector conversion of the second health and medical data of (A2) using the knowledge vector information (21); the converted vector information; a machine learning unit (74) that inputs the policy information (34) of the No. 3 database (DB#3) and outputs model information (41) that models the relationship between the vector information and the policy information (34) by machine learning; ) and a fourth database (DB#4) for managing the output model information (41), and the second group (A2) constitutes the first group (A1) and the policy information (34) includes the subjects of the health care policy in the second group (A2) and the result of the policy.

With the above configuration, the healthcare data analysis system calculates vector information 23 and knowledge vector information 21 from nationwide healthcare data, and calculates vector information 23, policy information 34 of a specific area A2, and healthcare data of area A2. Model information (prediction model) 41 is generated from the data. By applying the healthcare data of another region A3 (or country A1) to this model information 41, predicting the effect when the measures of a specific region A2 are implemented in another region A3 (or country A1) becomes possible. This makes it possible to support creation of new measures for health care (medical care, physical examination, or nursing care).

(2) The health and medical data analysis system according to (1) above, which manages the third health and medical data (health checkup information 51, receipt information 52, nursing care information 53) of the third group (A3). a fifth database (DB#5); a second vector conversion unit (84) for generating vector information by vector conversion of the third health care data using the knowledge vector information (21); further comprising: a first estimation calculation unit (85) for applying the converted vector information to the model information (41) to perform a simulation; and a first output unit (83) for outputting the result of the simulation. and the third group (A3) is a subgroup constituting the first group (A1) and is composed of a group different from the second group (A2), and the first estimation operation A health and medical care data analysis system characterized in that a part (85) estimates a result of the health care measures implemented in the second group (A2) being executed in the third group (A3).

With the above configuration, the healthcare data analysis system can generate model information (prediction model) from the vector information 23, the policy information 34 of the specific area A2 (second group), and the healthcare data of the area A2 (second group). 41 is generated. By applying the healthcare data of another area A3 (third group) to this model information 41, the measures of a specific area A2 (second group) are implemented in another area A3 (third group). It is possible to predict the effect of

(3) In the health care data analysis system described in (1) above, a third vector conversion for generating vector information by vector conversion of the first health care data using the knowledge vector information (21) a second estimation calculation unit (62) for performing a simulation by applying the vector-transformed vector information to the model information (41); and a second output unit for outputting the result of the simulation. and (63), wherein the second estimation calculation unit (65) executes the health care measures implemented in the second group (A2) in the first group (A1) A healthcare data analysis system characterized by estimating results.

With the above configuration, the healthcare data analysis system can generate model information (prediction model) from the vector information 23, the policy information 34 of the specific area A2 (second group), and the healthcare data of the area A2 (second group). 41 is generated. By applying the health care data of country A1 (first group) to this model information 41, the effect of implementing measures in a specific region A2 (second group) in country A1 (first group) can be predicted.

(4) The health and medical data analysis system according to (1) above, wherein the first health and medical data of the first group (A1) is anonymous information. system.

With the above configuration, personal information can be protected in national level health and medical data by concealing personal information and using the data.

(5) In the health care data analysis system described in (2) above, the model information (41) of the fourth database (DB#4) includes an identifier of the policy information (34), An estimation calculation unit (85) of No. 1 outputs results of measures for each of a plurality of model information (41), receives selection of the model information (41), and calculates a second model information (41) based on the selected model information (41). A health care data analysis system characterized by performing a simulation of health care data of 3 populations (A3).

With the above configuration, the user of the healthcare data analysis system displays the results of measures for each of the plurality of model information (41) on the output device, selects the desired model information (41), and selects the selected model. Information 41 allows simulation of healthcare data for another region A3 (third population).

(6) In the health care data analysis system described in (3) above, the model information (41) of the fourth database (DB#4) includes an identifier of the policy information (34), 2, the estimation calculation unit (65) outputs the results of the measures for each of the plurality of model information (41), receives the selection of the model information (41), and uses the selected model information (41) A healthcare data analysis system characterized by executing a simulation.

With the above configuration, the user of the healthcare data analysis system displays the results of measures for each of the plurality of model information (41) on the output device, selects the desired model information (41), and selects the selected model. Information 41 may simulate healthcare data for country A1 (first population).

In addition, the present invention is not limited to the above-described embodiments, and includes various modifications. For example, the above embodiments are described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the described configurations. In addition, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. In addition, addition, deletion, or replacement of other configurations for a part of the configuration of each embodiment can be applied singly or in combination.

Further, each of the above configurations, functions, processing units, processing means, and the like may be realized by hardware, for example, by designing them in an integrated circuit. Further, each of the above configurations, functions, etc. may be realized by software by a processor interpreting and executing a program for realizing each function. Information such as programs, tables, and files that implement each function can be stored in recording devices such as memories, hard disks, SSDs (Solid State Drives), or recording media such as IC cards, SD cards, and DVDs.

Further, the control lines and information lines indicate those considered necessary for explanation, and not all control lines and information lines are necessarily indicated on the product. In practice, it may be considered that almost all configurations are interconnected.

1, 2, 3, 4, 5 Databases 6, 8 Simulation device 7 Model generation device 9 Knowledge creation device 11, 31, 51 Health checkup information 12, 32, 52 Receipt information 13, 33, 53 Nursing care information 21 Knowledge vector information 23 Vector information 34 Policy information 41 Model information 61, 71, 81, 91 CPU
52, 62, 72, 82, 82 memories 63, 83 output units 64, 84 vector conversion units 65, 82 estimation calculation unit 93 knowledge vector generation unit

Claims (12)

A computer system having a processor and memory is a health data analysis system for analyzing health data,
a first database managing first health data for a first population;
a knowledge vector creation unit that inputs the first health and medical data, converts the mutual relationship between the first health and medical data into a vector, and outputs knowledge vector information;
a second database that stores the knowledge vector information;
a third database that manages second health and medical data of a second group and policy information about policies implemented in the second group;
inputting the second health care data of the third database and the knowledge vector information of the second database, and performing vector conversion of the second health care data of the second group using the knowledge vector information a first vector conversion unit that generates vector information;
a machine learning unit that inputs the converted vector information and the policy information of the third database, and outputs model information obtained by modeling the relationship between the vector information and the policy information by machine learning;
a fourth database that manages the output model information;
The second group is composed of subgroups that constitute the first group,
The health and medical care data analysis system, wherein the policy information includes subjects of the health and medical care policy in the second group and results of the policy. A health care data analysis system according to claim 1,
a fifth database managing third health data for a third population;
a second vector conversion unit that generates vector information by vector conversion of the third health care data using the knowledge vector information;
a first estimation calculation unit that performs a simulation by applying the vector-transformed vector information to the model information;
a first output unit for outputting the simulation result;
The third group is
A subgroup that constitutes the first group and is composed of a group different from the second group,
The first estimation calculation unit
A health and medical care data analysis system, characterized by estimating the results of health and medical measures carried out by the second group in the third group. A health care data analysis system according to claim 1,
a third vector conversion unit that generates vector information by vector conversion of the first health and medical data using the knowledge vector information;
a second estimation calculation unit that performs a simulation by applying the vector-transformed vector information to the model information;
a second output unit for outputting the result of the simulation;
The second estimation calculation unit
A health and medical care data analysis system, characterized by estimating a result of health and medical measures implemented by the second group being executed by the first group. A health care data analysis system according to claim 1,
The health and medical data analysis system, wherein the first health and medical data of the first group is anonymous information. A health care data analysis system according to claim 2,
the model information of the fourth database includes an identifier of the policy information;
The first estimation calculation unit
Health and medical data characterized by outputting a result of measures for each of a plurality of model information, accepting selection of the model information, and performing a simulation of the third health and medical data using the selected model information. analysis system. A health care data analysis system according to claim 3,
the model information of the fourth database includes an identifier of the policy information;
The second estimation calculation unit
outputting results of measures for each of a plurality of said model information; receiving selection of said model information; and performing a simulation of said first health and medical data using said selected model information. Data analysis system. A computer having a processor and a memory is a health data analysis method for analyzing health data,
The calculator inputs the first health and medical data from a first database that manages the first health and medical data of the first group, and converts the mutual relationship between the first health and medical data into a vector. a knowledge vector creation step of converting and outputting as knowledge vector information and storing the knowledge vector information in a second database;
The computer inputs the second health and medical data from a third database that manages the second health and medical data of the second group and the policy information about the policy implemented in the second group. a first vector conversion step of inputting knowledge vector information from the second database and generating vector information by vector conversion of the second health care data of the second group using the knowledge vector information;
The computer inputs the converted vector information and the policy information of the third database, and stores model information obtained by modeling the relationship between the vector information and the policy information by machine learning in the fourth database. a machine learning step;
The second group is composed of subgroups that constitute the first group,
The health and medical care data analysis method, wherein the policy information includes subjects of the health and medical care policy in the second group and results of the policy. The health care data analysis method according to claim 7,
The computer inputs the third health and medical data from a fifth database managing third health and medical data of a third population, inputs knowledge vector information from the second database, and inputs the knowledge vector information from the second database. a second vector conversion step of generating vector information by vector conversion of the health care data using the knowledge vector information;
a first estimation calculation step in which the computer performs a simulation by applying the vector information obtained by the vector conversion to the model information;
a first output step in which the calculator outputs the results of the simulation;
The third group is
A subgroup that constitutes the first group and is composed of a group different from the second group,
The first estimation calculation step includes:
A method of analyzing health and medical care data, characterized by estimating a result of the health and medical care measures carried out by the second group being carried out by the third group. The health care data analysis method according to claim 7,
a third vector conversion step in which the calculator generates vector information by vector conversion of the first health care data using the knowledge vector information;
a second estimation calculation step in which the computer performs a simulation by applying the vector information obtained by vector conversion to the model information;
a second output step in which the computer outputs the results of the simulation;
The second estimation calculation step includes:
A method of analyzing health and medical care data, characterized by estimating a result of implementing health and medical care measures in the second group in the first group. The health care data analysis method according to claim 7,
The health and medical data analysis method, wherein the first health and medical data of the first group is anonymous information. The health care data analysis method according to claim 8,
the model information of the fourth database includes an identifier of the policy information;
The first estimation calculation step includes:
outputting results of measures for each of a plurality of said model information; accepting selection of said model information; and performing a simulation of said third health and medical data using said selected model information. Data analysis method. The health care data analysis method according to claim 9,
the model information of the fourth database includes an identifier of the policy information;
The second estimation calculation step includes:
outputting results of measures for each of a plurality of said model information; receiving selection of said model information; and performing a simulation of said first health and medical data using said selected model information. Data analysis method.

Images