JP2018180993A - Data analysis support system and data analysis support method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform an analysis in consideration of the cause of occurrence of an event to be predicted. SOLUTION: The data analysis support system is a storage unit that holds medical information regarding the states of a plurality of patients, and patient state classification that classifies the states of the plurality of patients into a plurality of clusters based on the medical information. A prediction model that predicts the occurrence of the predetermined event from the condition of the classified patient based on the unit, the condition of the classified patient, and whether or not a predetermined event has occurred in each of the patients. It has a prediction model generation unit to be generated. [Selection diagram] Fig. 1

Description

  The present invention relates to an information system that supports analysis of accumulated data in a medical institution.

  In recent years, by utilizing accumulated data in medical institutions, machine learning or statistical analysis is used to carry out prediction / verification of a specific event, etc. It is

  Patent Document 1 uses a prediction model created for a patient who is released (or will be discharged) from intensive monitoring such as ICU to an unmonitored environment, and the physiology of the patient's heart rate, body temperature, etc. It is described that the patient's death / re-hospitalization risk is calculated by utilizing physical information and treatment data, etc., and the discharge decision by the doctor is supported by presenting it to the doctor.

Japanese Patent Application Publication No. 2014-520335

  As mentioned above, in the treatment planning decision support system so far, the prediction model is created, and the created prediction model is used, and physiological information such as heart rate and temperature of the patient and treatment data are used to There has been a system for realizing support for discharge judgment by a doctor by calculating the occurrence risk of an event to be predicted and presenting it to a doctor. However, when creating a prediction model, there is a problem that the prediction model is not constructed in consideration of the cause of occurrence of the event to be predicted. For example, in the case of using only two values of whether or not to undergo rehospitalization as an objective variable used to create a prediction model when constructing a prediction model for rehospitalization, an explanatory variable related to each cause of rehospitalization However, it is not easy to grasp the cause when it is different, and it is difficult to determine an appropriate treatment plan to prevent a re-hospitalization, to make an appropriate discharge decision, and so on. Therefore, an analysis that takes into consideration the cause of occurrence of the event to be predicted has been a challenge in order to support decision making on treatment planning and support for appropriate discharge decision.

  According to the technology described in Patent Document 1, since the cause of occurrence of the event to be predicted is not considered, analysis in consideration of the cause of occurrence can not be performed.

  As described above, it has been difficult to obtain sufficient effects with the conventionally disclosed technology in order to realize analysis that takes into consideration the cause of occurrence of the event to be predicted.

  In order to solve the above problems, a storage unit for holding medical treatment information on a plurality of patient states, a patient state classification unit for classifying the plurality of patient states into a plurality of clusters based on the medical treatment information, and the classification Model generation for generating a prediction model that predicts the occurrence of the predetermined event from the state of the classified patient based on the status of the determined patient and whether or not the predetermined event has occurred in each of the patients Providing a data analysis support system characterized by comprising:

  According to one aspect of the present invention, the patient to be analyzed is classified by classifying the patient based on the patient state after the occurrence of the event to be predicted, and constructing a prediction model using the objective variable created based on the classification. It becomes possible to predict / estimate the occurrence risk of the event to be predicted and the state of the patient to be analyzed after the occurrence of the event to be predicted, and easily analyze the cause of the event to be predicted. can do. Furthermore, for example, by constructing a prediction model using an objective variable created based on classification, it is possible to easily grasp factors related to the state after the occurrence of the event to be predicted.

  Problems, configurations, and effects other than the above are clarified by the following description of the embodiments.

It is a block diagram which shows the structure of the treatment plan determination assistance system of Example 1 of this invention. It is explanatory drawing which shows the example of the patient information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the test | inspection information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the prescription information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the event information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the event cause information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the objective variable information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the objective variable information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the objective variable information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is explanatory drawing which shows the example of the event master information table which the treatment plan determination assistance system of Example 1 of this invention hold | maintains. It is a flowchart which shows operation | movement of the treatment plan determination assistance system of Example 1 of this invention. In Example 1 of this invention, an event data production | generation part is a flowchart which shows the operation | movement which performs the analysis process which produces | generates event data. In Example 1 of this invention, a patient classification | category part is a flowchart which shows operation | movement which performs the analysis process which classify | categorizes a patient state. It is explanatory drawing which shows the example of the screen for analysis condition reception which the treatment plan determination assistance system of Example 1 of this invention displays. The treatment plan determination support system of Example 1 of this invention WHEREIN: It is explanatory drawing which shows the example of the presentation screen of the analysis result which a display screen production | generation part produces | generates, when cause information exists. The treatment plan determination assistance system of Example 1 of this invention WHEREIN: It is explanatory drawing which shows the example of the presentation screen of the analysis result which a display screen production | generation part produces | generates, when cause information does not exist. The treatment plan determination assistance system of Example 1 of this invention WHEREIN: When a clustering result display button is pressed down, it is explanatory drawing which shows the example of the display screen of the clustering result which a display screen production | generation part produces | generates. Description of an example of a screen of a clustering result generated by the display screen generation unit when the treatment plan determination support system of the first embodiment of the present invention implements clustering so that the value of a variable becomes high in prediction model construction FIG. In the treatment plan determination support system of Example 1 of this invention, it is explanatory drawing which shows the example of the display screen of the clustering result after performing weighting by a variable factor.

  Hereinafter, embodiments of the present invention will be described based on the drawings. The embodiments of the present invention are not limited to the embodiments described later, and various modifications can be made within the scope of the technical idea thereof.

«Configuration»
<Description of system configuration 1>
FIG. 1 is a block diagram showing the configuration of a treatment plan determination support system 100 according to a first embodiment of the present invention.

  The treatment plan determination support system of this embodiment is a computer system configured of a treatment plan determination support system 100, a hospital information system 120, an input / output terminal 130, and a network 140. The treatment plan determination support system of this embodiment creates analysis data from the accumulated data, and among the created analysis data, the patient status after the occurrence of the event to be analyzed is the patient state after the occurrence of the event to be predicted. Based on the classification of the patient based on the classification of the patient and the construction of the prediction model using the objective variable created based on the classification, the occurrence risk of the target event of the analysis target patient and the occurrence of the target event of the analysis target patient Later, it predicts and estimates what kind of state it is, and visualizes and outputs the result.

  In this embodiment, the input / output terminal 130 communicates with an input unit (not shown) such as a keyboard, a mouse or a touch panel, an output unit (not shown) such as a display, and the treatment plan determination support system 100 etc. (Not shown) and one or more personal computers. In addition, a PDA, a PHS, a mobile phone, a smartphone, a tablet terminal, and a glass including an input unit such as a button or a touch panel, an output unit such as a display, and a communication unit for communicating with the treatment plan determination support system 100 or the like. A portable terminal such as a wearable display terminal such as a flat panel display or a head mounted display may also be used as the input / output terminal 130.

  In this embodiment, the input / output terminal 130 is installed and used in a medical institution (health care provider) such as a hospital or a clinic. On the other hand, a treatment plan determination support system 100 is installed in the data center.

  Thus, by installing the treatment plan determination support system 100 at the data center, it is possible to centrally manage privacy information such as patient's personal information and data collected from the patient, thus simplifying security management such as information leakage prevention. it can. The treatment plan determination support system 100 may be installed and used in a healthcare provider depending on the type of operation.

  As a user of the input / output terminal 130 (hereinafter referred to as a user), a doctor at a medical institution, a person in charge of analysis, a pharmacist, a manager or a manager is assumed.

  The user operates the input / output terminal 130, creates analysis data from the accumulated data using the information system shown in the present embodiment, and among the created analysis data, a patient for which an event to be analyzed has occurred Classification of patients based on patient status after occurrence of predicted events. And the information system of a present Example builds the prediction model using the objective variable created based on classification, The occurrence risk etc. of the prediction object event of the analysis object patient, and the analysis object patient predicts It predicts and estimates what kind of state it is after the occurrence of the target event and visualizes the result.

  The treatment plan determination support system 100 includes a control unit 101, an output unit 102, a memory 103, a communication unit 104, a display screen generation unit 105, a data extraction unit 106, an analysis target patient extraction unit 107, and event data generation connected mutually. The unit 108 includes a patient state classification unit 109, an objective variable label generation unit 110, a prediction model generation unit 111, an integrated database 112, an event information storage database 113, a prediction model storage database 114, and an accumulated data acquisition unit 119.

  The accumulated data acquisition unit 119 is data of medical care information on the patient such as patient information, examination information, prescription information, treatment information, etc. accumulated in the medical care information database 121 present in the hospital information system 120 installed in the health care provider. Are stored in the integrated database 112.

  The accumulated data acquisition unit 119 may be activated directly by the user, or may be activated automatically at a time specified by the user in advance, for example, every Saturday night. Alternatively, the data may be automatically activated when the data of the medical care information database 121 is updated.

  The control unit 101 is, for example, a processor that executes a program stored in the memory 103, and controls each unit of the treatment plan determination support system 100. The memory 103 is, for example, a storage device such as a dynamic random access memory (DRAM), and stores data (for example, a program executed by the control unit 101) referred to by each unit of the treatment plan determination support system 100.

  The integrated database 112 may be stored in the memory 103 or may be stored in another storage device (for example, a non-volatile storage device such as an HDD (Hard Disk Drive), not shown) in the treatment plan determination support system 100 It is also good.

  The output unit 102 is a device that outputs the result of processing by the treatment plan determination support system 100, and may be, for example, a display device.

  The communication unit 104 is connected to the network 140 and communicates with the input / output terminal 130.

  The display screen generation unit 105, the data extraction unit 106, the analysis target patient extraction unit 107, the event data generation unit 108, the patient state classification unit 109, the objective variable label generation unit 110, and the prediction model generation unit 111 A processing unit that executes processing for realizing the functions of 100, and may be realized by dedicated hardware or software. In the latter case, the processing executed by each processing unit described above in the following description is actually executed by the control unit 101 according to an instruction described in a program stored in the memory 103. Details of the process executed by each of the above processing units will be described later.

  The program executed by the control unit 101 is provided to the treatment plan determination support system 100 via removable media (CD-ROM, flash memory, etc.) or the network 140, and stored in a non-volatile storage device which is a non-transitory storage medium. Ru. For this reason, the treatment plan determination support system 100 may have an interface (not shown) that reads data from removable media.

  The treatment planning decision support system 100 is a computer system configured on physically one computer, or on a plurality of logically or physically configured computers, and is performed by separate threads on the same computer. It may operate, and may operate on a virtual computer built on a plurality of physical computer resources.

  The hospital information system 120 is a computer system connected to the network 140, and may have, for example, a control unit, a memory, a communication unit (all not shown) and the like similar to those of the treatment plan determination support system 100. Good. The medical care information database 121 may be stored in the memory of the hospital information system 120, or may be stored in another storage device (not shown) in the hospital information system 120, such as a hard disk drive (HDD). Good.

  A treatment planning decision support system 100, a hospital information system 120 and an input / output terminal 130 are connected to the network 140. The treatment planning decision support system 100 communicates with the hospital information system 120 and the input / output terminal 130 via the network 140.

  The network 140 uses wired communication with a LAN (Local Area Network) cable or wireless communication with a wireless LAN or the like.

  The network 140 can also use other wide area networks such as the Internet, VPN, a cellular phone communication network, a PHS communication network, and the like.

  Subsequently, a table structure constituting the integrated database 112 will be described.

  The integrated database 112 includes a patient information table 200 for managing basic information and the like for each patient, an examination information table 300 for storing examination information for each patient, and a prescription information table 400 for storing prescription information for each patient. . The integrated database 112 may also have a table (not shown) storing medical treatment information on patients such as treatment information for each patient.

  FIG. 2 is an explanatory view showing an example of a patient information table 200 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The patient information table 200 includes a field 201 for storing a record ID for identifying a record in the patient information table 200, a field 202 for storing a patient ID for identifying a patient, a field 203 for storing gender of a patient, and A field 204 for storing the patient's age at the time of admission to the medical institution or the date of visit to the medical institution for outpatient diagnosis, and the date of diagnosis when the patient visits the medical institution for outpatient diagnosis A field 205 for storing, a field 206 for storing the hospitalization date when the patient is admitted to the medical institution, and a field 207 for storing the discharge date when the patient is admitted to the medical institution and the hospital is withdrawn. And one or more fields (in the example shown in FIG. That field 208, a field 210) for storing the fields 209 and Disease 3 stores disease name 2, in constructed.

  When a patient visits a medical institution for outpatient diagnosis, the field 205 for storing the date of diagnosis stores a value for each record, and the field 206 for storing the date of hospitalization and the date of discharge "0" is stored in 207. On the other hand, when the patient visits a medical institution for admission to hospital, "0" is stored in the field 205 for storing the date of diagnosis, and the field 206 for storing the date of hospitalization and the date of discharge A value for each record is stored in 207.

  For example, an example 200A of the patient information record of the patient information table 200 shown in FIG. 2 is a record including information on the patient identified by the patient ID “# 1” and identified by the record ID “1”. The values for each field are gender "male", age "68", diagnosis date "0", hospitalization date "2016/04/01", discharge date "2016 Apr. 26", disease name 1 "Heart failure", disease 2 "renal failure", disease 3 is "NULL" indicating that no information is registered. This indicates that a 68-year-old male patient identified by patient ID "# 1" was admitted from April 1, 2016 to April 26, 2016 for heart failure and renal failure.

  The patient information table 200 may include information indicating four or more disease names (not shown), and a code indicating a disease name such as an ICD (International Classification of Diseases) code is stored instead of the disease name. It may be (not shown).

  FIG. 3 is an explanatory view showing an example of the examination information table 300 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The examination information table 300 is a table for storing information on an examination performed on a patient, and stores a field 301 for storing a patient ID, a field 302 for storing information for identifying an examination item, and an examination date. It comprises a field 303 and a field 304 for storing inspection results (for example, inspection values).

  For example, in the example of the examination information table 300 shown in FIG. 3, the value of the examination item “examination item X” is “41”, “2016/01/01” in the patient of the patient ID “# 1”. 04/07, "62", "2016/04/13", "180", "2016/04/15", "220", "2016/04/18", "196", "2016/04 / It shows that it was "120" to 25 ", and has shown that the value of the inspection item" inspection item Y "was" 38 "to" 2016/04/1 ".

  The inspection value stored in the field 304 described above is an example of the inspection result, and the inspection information table 300 is an image inspection (for example, CT image inspection) instead of (or in addition to) the inspection value. , And may include information indicating test results of types other than test values, and may further include information on patient's subjective symptoms such as "nausea" or "vomiting" (not shown).

  FIG. 4 is an explanatory view showing an example of a prescription information table 400 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The prescription information table 400 is a table for storing information on the prescription of a medicine given to a patient, and includes a field 401 for storing a patient ID, and a field 402 for storing information (drug name) for identifying a medicine. A field 403 stores the prescription start date of each drug, and a field 404 stores the prescription end date of each drug.

  For example, in the example of the prescription information table 400 shown in FIG. 5, in the patient with the patient ID "# 1", the drug with the drug name "drug A" is from "2016 Apr. 01" to "2016 Apr. 07" And, it is prescribed in the period from "2014/06/08" to "2016/04/14", and the drug of the drug name "drug B" is from "2016/04/03" to "2016/04/08" The drug of the drug name "drug D" is prescribed during the period from "2016/04/16" to "2016/04/20". Further, the prescription information table 400 may have fields (not shown) for storing information on start time, end time, prescription, dosage, and efficacy of prescription of each medicine.

  Subsequently, a table structure constituting the event information storage database 113 will be described.

  The event information storage database 113 is configured by an event information table 500, an event cause information table 600, objective variable information tables 700 to 900, and an event master information table 1000. The event information table 500 stores information such as whether an event to be predicted has occurred in a specific patient. The event cause information table 600 stores information on the cause of occurrence of the event to be predicted. The objective variable information table 700 stores the classification result of the patient state classification unit 109 and the objective variable label created by the objective variable label generation unit 110 when the cause of occurrence of the event to be predicted is present. Here, the case where the cause of occurrence of the event to be predicted is present is the case where it is determined in step 1302 of FIG. 13 described later that the information related to the cause of the event exists. The objective variable information table 800 stores the classification result of the patient state classification unit 109 and the objective variable label created by the objective variable label generation unit 110 when there is no cause of occurrence of the event to be predicted. The objective variable information table 900 stores the classification result of the patient state classification unit 109 and the objective variable label created by the objective variable label generation unit 110 in the patient state when there is a plurality of pieces of patient classification information. The event master information table 1000 stores information of a prediction model corresponding to the details of event information generated by the event data generation unit.

  FIG. 5 is an explanatory view showing an example of the event information table 500 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The event information table 500 includes a field 501 storing an identifier of each record of the event information table, a field 502 storing an event identifier corresponding to each record, a field 503 storing a patient ID corresponding to each record, and the like. A field 504 stores an event occurrence date corresponding to a record, and a field 505 stores an event value corresponding to each record.

  When the event identified by the event ID corresponding to each record occurs in the patient identified by the patient ID corresponding to each record, “1” is stored in the field 505 storing the event value of the record If not occurred, "0" is stored.

  FIG. 6 is an explanatory view showing an example of the event cause information table 600 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The event cause information table 600 includes a field 601 for storing an identifier of each record of the event cause information table, and a field 602 for storing an occurrence cause of an event corresponding to each record.

  In the field storing the cause of occurrence of an event, for example, the cause of occurrence of an event corresponding to each event record ID, such as “hypertension control failure” or “arrhythmia” is stored. A number or an identifier indicating the cause of the event may be stored.

  FIG. 7 is an explanatory view showing an example of the target variable information table 700 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The objective variable information table 700 includes a field 701 for storing an identifier of each record of the event cause information table, a field 702 for storing an event value corresponding to each record, and a field for storing a value of a patient cluster corresponding to each record. A field 703 stores the value of the objective variable label corresponding to each record.

  In the field storing the value of the objective variable label, the value corresponding to each patient cluster is stored. For example, when "hypertension control failure" is stored as the value of the patient cluster, the value of "A" is stored as the objective variable label, and when "arrhythmia" is stored as the value of the patient cluster, the objective variable label The value of "B" is stored as When “0” is stored as an event value, “Null” indicating that an event has not occurred as a patient cluster is stored, and “Z” is stored as a target variable label.

  FIG. 8 is an explanatory view showing an example of the target variable information table 800 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The objective variable information table 800 includes a field 801 for storing an identifier of each record of the event cause information table, a field 802 for storing an event value corresponding to each record, and a field for storing a value of a patient cluster corresponding to each record. A field 803 stores the value of the objective variable label corresponding to each record.

  In the field 804 for storing the value of the objective variable label, a value corresponding to each patient cluster calculated by the clustering process is stored. For example, when "cluster 1" is stored as the value of patient cluster, the value of "A" is stored as the target variable label, and when "cluster 2" is stored as the value of patient cluster, the target variable label The value of "B" is stored as When “0” is stored as an event value, “Null” indicating that an event has not occurred as a patient cluster is stored, and “Z” is stored as a target variable label.

  FIG. 9 is an explanatory view showing an example of the target variable information table 900 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The objective variable information table 900 stores a field 901 storing an identifier of each record of the event cause information table, a field 902 storing an event value corresponding to each record, and a value of the patient cluster 1 corresponding to each record. It comprises a field 903, a field 904 storing the value of the patient cluster 2 corresponding to each record, and a field 905 storing the value of the objective variable label corresponding to each record.

  The field 903 storing the value of patient cluster 1 stores the clustering result of the patient based on the cause information, and the field 904 storing the value of patient cluster 2 corresponds to each patient cluster calculated by the clustering process. The value to be stored is stored.

  A value corresponding to the patient cluster 2 is stored in the field 905 for storing the value of the objective variable label. For example, when "cluster 1 '" is stored as the value of the patient cluster 2, the value of "A'" is stored as the objective variable label. In addition, when “0” is stored as the event value, “Null” indicating that the event has not occurred as patient cluster 1 and patient cluster 2 is stored, and “Z” is stored as the target variable label .

  FIG. 10 is an explanatory view showing an example of the event master information table 1000 held by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The event master information table 1000 includes a field 1001 for storing an event identifier corresponding to each record, a field 1002 for storing a condition 1 at the time of creation of a prediction model corresponding to the event identifier, and creation of a prediction model corresponding to the event identifier. A field 1003 for storing the condition 2 at the time, a field 1004 for storing an objective variable of the prediction model corresponding to each event identifier, and a field 1005 for storing an identifier of the prediction model corresponding to each event identifier . In the above example, the event master information table 1000 includes two conditions at the time of creation of the prediction model corresponding to the event identifier, but may include three or more conditions. (Not shown).

  Next, the operation flow of the system of this embodiment will be described using a flowchart.

  FIG. 11 is a flowchart showing the operation of the treatment plan determination support system 100 according to the first embodiment of the present invention.

  First, the control unit 101 executes step (S) 1101 and calls patient data accumulated from the integrated database 112. Here, data on patients, for example, information for identifying a patient, information on prescription drugs corresponding to a patient, information on examination results corresponding to a patient, etc. are collectively referred to as patient data. At step 1101, the control unit 101 may read out all patient data included in the integrated database 112 and store it in the memory 103.

  Note that this step is executed in accordance with an instruction described in a program stored in the memory 103. The other steps described below are the same.

  Next, the control unit 101 executes step 1102, displays an acceptance screen of analysis conditions, and receives information on a patient to be analyzed and information on analysis conditions.

  FIG. 14 is an explanatory view showing an example of an analysis condition acceptance screen displayed by the treatment plan determination support system 100 according to the first embodiment of the present invention.

  An example of the screen for receiving analysis conditions 1400 is a selection area 1401 of a patient to be analyzed, a setting area 1402 of a use model, a specification condition area 1403 of a patient for analysis data, a setting area 1404 of objective variables, a statistical analysis option And an analysis execution button 1406.

  The user selects a patient to be analyzed regarding the occurrence of an event as a patient to be analyzed from among the patients displayed in the selection area 1401 of the patient to be analyzed.

  The user operates the radio button attached to the setting area 1402 of the use model to select either “Create new model” or “existing model”. If the former is selected, a predictive model for the occurrence of an event is newly created, and that model is used for analysis on the occurrence of the event. If the latter is selected, an existing model created in past analysis or stored as data in advance is used for analysis.

  When the user selects “Create a new model”, a designated condition area 1403 of a patient for analysis data, a setting area 1404 of a target variable, and a setting area 1405 of a statistical analysis option are displayed. The user operates the pull-down list in the analysis data patient specification condition area 1403 to set the analysis data patient specification conditions used to create a new model. In the example of FIG. 14, “disease name = heart failure” is selected as condition 1 and “age> = 60 years old” is selected as condition 2. This indicates that a patient with a heart failure disease name and an age of 60 or more is selected as a patient for analysis data. The analysis data patient designation condition area 1403 may include a pull-down list for selecting a third or more condition (not shown).

  Also, the user operates the pull-down list in the target variable setting area 1404 to set a target variable used for creating a new model. In the example of FIG. 14, “re-admission within 30 days after discharge from hospital” is selected as the condition 1. This indicates that information indicating whether a re-hospitalization within 30 days after discharge has occurred is used as a target variable. The target variable setting area 1404 may include a pull-down list for selecting a second or more condition (not shown).

  When the user selects the “existing model”, the control unit 101 reads out the conditions and objective variables of the prediction model created in the past stored in the event master information table 1000, and uses them from the model created in the past The screen for making the user select a model to be displayed is displayed (not shown).

  In addition, the user operates the pull-down list in the setting area 1405 of the statistical analysis option to set the statistical method used for creating a new model. In the example of FIG. 14, “logistic regression” is selected as the statistical method, and “AIC: 100” is selected as the variable selection. This indicates that construction of a prediction model is performed using logistic regression, utilizing 100 variables selected by Akaike's Information Criterion (AIC).

  In the statistical analysis option setting area 1405, for example, “Support Vector Machine (SVM)”, “Deep-Learning”, “neural network”, “Bayesian network”, “decision tree learning”, “random forest”, etc. A method of machine learning may be presented, and one of them may be selected by the user, or a plurality of machine learning methods as in the above example may be selected, and a plurality of prediction models may be created simultaneously. A prediction model creation method (ensemble learning) in which a plurality of machine learning methods are combined may be selected (not shown).

  In addition, in the variable selection of the target variable setting area 1404, for example, any other method such as “stepwise”, “correlation analysis”, “maximum information coefficient” and “L1 regularization” may be selected. Alternatively, a plurality of variable selection methods may be selected and a plurality of variable selection methods may be simultaneously selected as in the above example, or a plurality of variable selection may be performed a plurality of times. You may choose the method to perform (illustration omitted).

  When the user selects the analysis execution button 1406, the control unit 101 executes step 1103, activates the event data generation unit 108, and based on the various analysis conditions received in step 1102, the patient for analysis data set in step 1102 Event data is created using the setting information of and the setting information of the target variable. The details of the process executed in step 1103 will be described with reference to FIG.

  FIG. 12 is a flowchart showing an operation of the event data generation unit 108 executing analysis processing for generating event data in the first embodiment of the present invention.

  At step 1201, the event data generation unit 108 reads out the patient information table 200 onto the memory 103.

  Next, in step 1202, the event data generation unit 108 sets “0” in the event record ID.

  Next, in step 1203, the event data generation unit 108 acquires all the record IDs of the patient information table 200 read out to the memory 103, and determines that the event data generation process is to be executed for any record ID. Step 1204 is performed. Specifically, for example, the event data generation unit 108 is in the event data generation process (that is, the process after step 1204) among all the records of the patient information table 200 corresponding to the record ID acquired in step 1203. Execute event data generation processing for unexecuted records.

  In step 1204, the event data generation unit 108 determines whether the patient of the record ID determined to execute event data generation processing in step 1203 matches the condition received in step 1102. If it is determined that they match, the event data generation unit 108 executes the next step 1205.

  In step 1205, the event data generation unit 108 sets “0” as the event value, and adds “1” to the event record ID.

  Next, in step 1206, the event data generation unit 108 determines in the event information table 500 that the event record ID obtained by the addition in step 1205, the event ID, and event data generation processing in step 1203. The patient ID of the record ID and the event value are linked and stored.

  Next, in step 1207, if it is determined that the data matches the information of the target variable received in step 1102, the event data generation unit 108 executes the next step 1208.

  Next, in step 1208, the event data generation unit 108 sets the event value of the record storing the data in step 1206 to 1, and stores the date matched with the information with the objective variable as the event occurrence date.

  For example, as shown in FIG. 14, condition 1 “disease name = heart failure” and condition 2 “age> = 60 years” are selected as analysis conditions, and condition 1 “re-hospitalization within 30 days after discharge” as the objective variable. Is selected, the event data generation unit 108 determines whether each record of the patient information table 200 matches the analysis condition, and determines whether the condition of the objective variable is satisfied for the matching record.

  For example, when the top record 200A of the patient information table 200 shown in FIG. 2 is selected as the target of processing in step 1203, the patient corresponding to the record 200A is 60 years or older, and the corresponding disease name includes heart failure. Therefore, it is determined in step 1204 that the record 200A meets the analysis condition. Then, in step 1205, a new record (for example, record 500A) corresponding to the record 200A is added to the event information table 500. At this point, an event ID (in the above example, condition 1 “re-hospitalization within 30 days after discharge”) is identified in the field 502 of the event ID of the record 500A, which corresponds to the condition set as the objective variable For example, "event 1" is set, and "0" is set in the field 505 of the event value. Further, at step 1206, the same value as the patient ID of the record 200A is set in the field 503 of the patient ID of the record 500A.

  Thereafter, in step 1207, it is determined whether the record 200A matches the setting condition of the objective variable. In the example of FIG. 2, after the patient of record 200A is discharged on April 26, 2016, he is hospitalized again on May 18, 2016 (record 200B). That is, since the condition 1 “re-hospitalization within 30 days after discharge” is satisfied (step 1207: Yes), the field 505 of the event value of the record 500A is changed to “1” (step 1208). Are registered in the field 504 of the event occurrence date. In this example, the event “re-hospitalized on May 18, 2016” is treated as an analysis target event in the process of FIG. 13 described later.

  If the patient is not admitted again by May 26, 2016, the condition 1 “re-admission within 30 days after discharge” is not satisfied (Step 1207: No), Step 1208 is not executed Returns to step 1203 and the event value field 505 of the record 500A remains "0".

  When step 1208 is completed, the process returns to step 1203 and the processes after step 1204 are repeatedly executed until the process is completed for all the acquired records. When the event data generation process is completed for all the acquired records, the process of FIG. 12 is completed, and the processes after step 1104 are executed.

  In step 1104, the control unit 101 activates the patient state classification unit 109, and classifies the patient based on the cause of the occurrence of the event or the patient state after the occurrence of the event. The details of the process executed in step 1104 will be described with reference to FIG.

  FIG. 13 is a flowchart showing an operation of the patient state classification unit 109 executing an analysis process of classifying a patient state in the first embodiment of the present invention.

  In step 1301, the patient state classification unit 109 reads out the event information table 500 created in step 1103 on the memory 103.

  Next, in step 1302, the patient state classification unit 109 verifies whether the information related to the cause of the event exists in the medical information database 121 or the integrated database 112, and the information related to the cause of the event is If it is determined that it exists, step 1303 is executed.

  Next, in step 1303, the patient state classification unit 109 reads out information related to the cause of occurrence of the event on the memory 103, associates the event record ID of the event information table 500 with the cause of occurrence of the read out event, and stores them. Do. Thereafter, the patient state classification unit 109 executes step 1307 without executing steps 1304 to 1306 described later.

  On the other hand, if it is determined that there is no information related to the cause of occurrence of the event, the patient state classification unit 109 executes step 1304 without executing step 1303.

  Next, in step 1304, the patient state classification unit 109 extracts all patient IDs of records having an event ID to be analyzed and an event value of “1” from the event information table 500.

  Next, in step 1305, the patient state classification unit 109 extracts the medical care information immediately after the occurrence of the event to be analyzed regarding the patient corresponding to the patient ID extracted in step 1304. Here, immediately after the occurrence of the event to be analyzed is an example of a predetermined period based on the time of occurrence of the event to be analyzed, and for example, a predetermined time has elapsed since the occurrence of the event to be analyzed. It is a period of time to Further, the medical care information extracted here is information indicating the state of the patient identified by the patient ID extracted in step 1304, and is obtained from at least one of the patient information table 200, the examination information table 300 and the prescription information table 400. It may also include the extracted values for the patient.

  Next, in step 1306, the patient state classification unit 109 performs clustering processing such as cluster analysis targeting the values included in the medical care information extracted in step 1305, and the patient group of the patient ID extracted in step 1304 Classify into multiple clusters. In clustering processing, the number of clusters may use a value designated by the user in advance, or a cluster designated by the user as the upper limit cluster number may be exhaustively searched for clusters with the highest prediction accuracy of events. I see. For example, the patient state classification unit 109 may execute clustering multiple times while changing the number of clusters. The prediction model generation unit 111 generates a plurality of prediction models based on the results of clustering multiple times (step 1106 described later), compares the prediction accuracy of the occurrence of the event by each prediction model, and determines the highest prediction model. You may output it. The patient state classification unit 109 may determine the number of clusters at that time as the optimum number of clusters. Alternatively, the patient state classification unit 109 may use a method such as the x-means method that automatically optimizes the number of clusters (not shown). This makes it possible to construct a prediction model with high prediction accuracy.

  Once step 1303 or step 1306 above is performed, then step 1307 is performed. If step 1307 is executed following step 1303 (that is, it is determined in step 1302 that event cause information exists), the patient state classification unit 109 sets the event cause corresponding to each event record ID as a patient cluster. The event record ID, the event value, and the patient cluster are associated with each other and stored in the target variable information table 700. On the other hand, if step 1307 is executed after step 1306 (ie, it is determined in step 1302 that there is no event cause information), the patient status classification unit 109 uses the cluster information classified in step 1306. Then, the clustering result of the patient ID corresponding to each event record ID is set as a patient cluster corresponding to each event record ID, and the event record ID, the event value, and the patient cluster are associated with each other to obtain the objective variable information table. Store in 800.

  Next, in step 1308, the patient state classification unit 109 extracts all the patient IDs of the record having the event ID to be analyzed and the event value of “0” from the event information table 500.

  Next, in step 1309, the patient state classification unit 109 sets Null in the patient cluster, associates the event record ID, the event value, and the patient cluster with each record extracted in step 1308, and the purpose is The variable information table 700 or the target variable information table 800 is stored.

  The patient state classification unit 109 executes step 1303 to generate a cluster corresponding to the event cause, but if the number of patients belonging to any of the clusters is smaller than the predetermined lower limit value. Steps 1304 to 1306 may be performed by reducing the number of clusters. FIG. 9 described above shows an example of the result of clustering performed in this manner. In addition, as a result of executing step 1304 to step 1306, the patient state classification unit 109 reduces the number of clusters even if the number of patients belonging to any cluster is smaller than the predetermined lower limit value, and performs step 1304 to step 1306 again. Can be performed.

  As described above, when the cause information can be used, it is possible to perform appropriate clustering according to the cause. If the cause information can not be used, clustering can be performed using the medical care information to classify patient states with similar tendency into the same cluster. In addition, since the number of patients included in each cluster is maintained at or above a predetermined lower limit value, it is possible to secure the accuracy of construction of a prediction model described later.

  Next, in step 1105, the control unit 101 activates the objective variable label generating unit 110, and in the objective variable information table 700 or the objective variable information table 800, creates an objective variable label unique to each patient cluster, and each event An objective variable label corresponding to each patient cluster of the record ID is registered in the field 704 or 804 (not shown).

  Next, in step 1106, the control unit 101 receives the patient information of the analysis data patient received in step 1102, medical care information such as examination information and prescription information, and information of the objective variable information table 700 or the objective variable information table 800. Are used to construct a prediction model based on the method of creating a prediction model received in step 1102. Specifically, based on the state of the patient classified by the patient state classification unit 109 and whether or not the event to be predicted has occurred in each of the patient states, the prediction model generation unit 111 determines from the state of the patient. Construct a prediction model that predicts the occurrence of the event to be predicted. The construction of a prediction model according to the above-described generation method is known, and thus the detailed description thereof is omitted.

  Next, in step 1107, the control unit 101 applies the prediction model created in step 1106 to the analysis target patient accepted in step 1102, and the occurrence probability of the event to be predicted of the analysis target patient and each purpose Calculate the probability etc. corresponding to the variable label.

  Next, in step 1108, the control unit 101 activates the display screen generation unit 105, and based on the occurrence probability of the event to be predicted of the patient to be analyzed and the probability corresponding to each objective variable label calculated in step 1107 , Generate a presentation screen of analysis results to be presented to the user.

  Next, in step 1109, the control unit 101 causes the output unit 102 to output the presentation screen generated in step 1108.

  FIG. 15 is an explanatory diagram showing an example of a presentation screen 1500 of an analysis result generated by the display screen generation unit 105 when there is cause information in the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The analysis result presentation screen 1500 shows an event occurrence risk presentation area 1501 that presents the occurrence risk of the prediction target event of the analysis target patient, and an event occurrence cause that presents the probability for each occurrence cause of the prediction target event of the analysis target patient. And a probability presentation area 1502.

  In this manner, by displaying each event occurrence cause probability in the event occurrence cause probability presentation area 1502, the user can easily grasp what cause the analysis target patient causes the analysis target event to occur. It is possible to determine an appropriate treatment plan to prevent the occurrence of an event, to make an appropriate discharge decision, and the like.

  FIG. 16 is an explanatory diagram showing an example of a presentation screen 1600 of an analysis result generated by the display screen generation unit 105 when there is no cause information in the treatment plan determination support system 100 according to the first embodiment of the present invention.

  The analysis result presentation screen 1600 presents similar cases among patients in the patient cluster in which the analysis subject patient is classified and an event occurrence risk presentation area 1601 which presents the occurrence risk of an event to be predicted of the analysis subject patient. A similar case presentation area 1602 and a clustering result display button 1603 used when displaying the screen of the clustering result.

  In this manner, by presenting similar cases among the patients in the patient cluster in which the analysis target patients are classified in the similar case presentation area 1602, the user is able to determine how the analysis target patient generates an event. It is possible to determine if the patient is in a similar state as a normal patient, analyze similar patients, determine an appropriate treatment plan to prevent the occurrence of an event, and make an appropriate discharge judgment, etc. Is possible. The user presses the clustering result display button 1603 when referring to the result in which the analysis target patient is classified by cluster analysis. The screen presented at this time will be described with reference to FIG.

  FIG. 17 shows an example 1700 of a display screen of the clustering result generated by the display screen generation unit 105 when the clustering result display button 1603 is pressed in the treatment plan determination support system 100 according to the first embodiment of the present invention. FIG.

  The display screen 1700 of the clustering result includes an axis selection area 1701 and a clustering result display area 1702. First, the user operates the pull-down list displayed in the axis selection area 1701 to set the vertical axis and horizontal axis of the clustering result displayed in the clustering result display area 1702. In the clustering result display area 1702, the clustering result of the analysis target patient is displayed according to the vertical axis and the horizontal axis selected by the pull-down list of the axis selection area 1701. In the screen example 1700, the case of the analysis target patient is indicated by an open circle, and the cases of other patients are indicated by a black circle. In this example, cases with patient IDs # 1 and # 20 are presented as cases similar to the analysis target patient in cluster A which is a cluster to which the analysis target patient belongs. Thus, displaying the clustering result on an arbitrary axis enables the user to analyze in which items it is determined that a case similar to the patient to be analyzed is similar. This allows the user to grasp what kind of patient is in a similar state when an event occurs, and analyze a similar patient to determine an appropriate treatment plan to prevent the occurrence of the event. It is possible to make appropriate discharge decisions, etc.

  In addition, in the clustering in step 1306, the factor designated by the user or the factor stored in advance in the system is defined as a variable factor, and the weighting value such as the variable factor is the highest in construction of the prediction model in step 1106 The number of clusters may be defined as the number of used clusters, and clustering may be performed. Here, the variable factor is a factor that can be changed by the treatment action. In the prediction model for predicting the event occurrence risk, when the weighting factor of the variable factor is sufficiently high, it is possible to perform treatment for changing the variable factor in the direction of reducing the event occurrence risk. The details of such clustering will be described with reference to FIGS. 18 and 19.

  FIG. 18 is a result of clustering generated by the display screen generation unit 105 when the treatment plan determination support system 100 according to the first embodiment of the present invention performs clustering such that the value of a variable becomes high in prediction model construction. It is an explanatory view showing an example 1800 of a screen.

  The clustering result display screen 1800 includes an axis selection area 1801, a variable factor selection area 1802, a variable factor weighting execution button 1803, and a clustering result display area 1804. The user first operates the pull-down list displayed in the axis selection area 1801 to set the vertical axis and the horizontal axis of the clustering result displayed in the clustering result display area 1804. The clustering result of the analysis target patient is displayed in the clustering result display area 1804 according to the vertical axis and the horizontal axis selected by the pull-down list of the axis selection area 1801. In the screen example 1800, the event occurrence risk is selected on the vertical axis, the factor Y is selected on the horizontal axis, and an example in which the analysis target patient belongs to cluster A is shown. In this example, in cluster A, there are only two cases where the risk of occurrence of an event is lower than that of the analysis subject, and the value of factor Y in those cases is almost unchanged. In such a case, even if the factor Y is a variable factor, that is, a factor that can be improved by the treatment, the degree of the risk of an event occurring if the treatment is performed only by comparing the patient groups within the same cluster It is difficult to analyze whether it falls or not.

  In such a case, the user operates the variable factor selection area 1802 to select a variable factor, and presses a variable factor execution button 1803. The patient state classification unit 109 executes clustering with the number of clusters in which the weighting factor of the variable selected in the prediction model construction becomes high as the optimum number of clusters (step 1306).

  FIG. 19 is an explanatory view showing an example of a display screen of the clustering result after the weighting by the variable factor is performed in the treatment plan determination support system 100 of the embodiment 1 of the present invention.

  That is, FIG. 19 shows an example of the display screen 1800 of the clustering result after the user operates the variable factor selection area 1802 to select the variable factor and presses the variable factor weighting execution button 1803.

  The display screen 1800 shown in FIG. 19 shows the result of performing clustering again using the number of clusters in which the weighting factor of the selected factor is high. In this example, the patient to be analyzed belongs to cluster A, and in cluster A, it is possible to analyze that the case where the event occurrence risk is lower than that of the patient to be analyzed tends to lower the factor Y. From this, there is a risk that the case of the patient to be analyzed has an event similar to the cluster A group of patients, and the lower the factor Y, the lower the risk. Analysis can be conducted to reduce the risk of event occurrence.

  In weighting factors, the degree of weighting may be specified by the user, a pre-described value may be used, or a value that maximizes the prediction accuracy of an event may be used. In addition, when the number of clusters in which the weighting factor of the selected factor is high is regarded as the optimum number of clusters, in order to maintain a constant prediction accuracy, the event prediction accuracy may be added on the condition that it exceeds a specified threshold. .

  The weighting of variable factors as described above may be performed, for example, in the following procedure. The patient state classification unit 109 executes clustering several times while changing the number of clusters (steps 1304 to 1306). The prediction model generation unit 111 generates a plurality of prediction models based on the results of the plurality of times of clustering (step 1106). Here, the prediction model generation unit 111 acquires weighting coefficients for the occurrence risk of the event to be predicted for a plurality of factors included in the plurality of generated prediction models, and generates a prediction model having a larger weighting factor of the variable factor. Output as

  As described above, the treatment plan determination support system according to the present invention classifies patients based on the patient state after occurrence of the target event for the patient in which the target event is analyzed, and uses the objective variable created based on the classification. By constructing a forecasting model, it is possible to predict and estimate the occurrence risk of the event to be predicted of the patient to be analyzed and the state of the patient to be analyzed after the occurrence of the event to be predicted. By performing visualization, it is possible to determine an appropriate treatment plan to prevent the occurrence of an event, to make an appropriate discharge decision, and the like.

  The present invention is not limited to the embodiments described above, but includes various modifications. For example, the embodiments described above have been described in detail for better understanding of the present invention, and are not necessarily limited to those having all the configurations of the description.

  For example, the treatment plan determination support system 100 of the above embodiment is an example of a data analysis support system for supporting analysis of medical information, and the present invention is used to support analysis of information other than medical information. It can also be done. Specifically, for example, the data analysis support system holds history information of a plurality of individuals, classifies the states of the plurality of individuals into a plurality of clusters based on the history information, and the states of the classified individuals; Generating a prediction model that predicts the occurrence of a predetermined event from the state of the classified individual based on whether or not the predetermined event has occurred in each individual, and applying the prediction model to the state of the individual to be analyzed The occurrence of a predetermined event in the individual to be analyzed may be predicted.

  Here, the individual may be a patient as in the above-described embodiment, or may be, for example, any type of device. The history information may be medical care information as in the above-described embodiment, and may include, for example, the type of device, usage period, maintenance history, failure history, examination history, and the like. The predetermined event may be, for example, re-hospitalization within a predetermined period as in the above embodiment, or may be occurrence of a failure of a predetermined mode.

  Further, each of the configurations, functions, processing units, processing means, etc. described above may be realized by hardware, for example, by designing part or all of them with an integrated circuit. Further, each configuration, function, and the like described above may be realized by software by the processor interpreting and executing a program that realizes each function. Information such as programs, tables, and files that realize each function is memory, hard disk drive, storage device such as SSD (Solid State Drive), or computer readable non-temporary data such as IC card, SD card, DVD, etc. It can be stored in a storage medium.

  Also, the drawings show control lines and information lines that are considered to be necessary to explain the embodiment, and not necessarily all control lines and information lines included in an actual product to which the present invention is applied. Not necessarily. In practice, almost all configurations may be considered to be mutually connected.

100 treatment plan determination support system 101 control unit 102 output unit 103 memory 104 communication unit 105 display screen generation unit 106 data extraction unit 107 analysis target patient extraction unit 108 event data generation unit 109 patient state classification unit 110 objective variable label generation unit 111 prediction Model generation unit 112 Integrated database 113 Event information storage database 114 Prediction model storage database 119 Accumulated data acquisition unit 120 Hospital information system 121 Medical care information database 130 Input / output terminal 140 Network

Claims (15)

A storage unit that holds medical treatment information on the states of a plurality of patients;
A patient state classification unit that classifies the states of the plurality of patients into a plurality of clusters based on the medical care information;
A prediction model is generated that predicts occurrence of the predetermined event from the state of the classified patient based on the classified patient state and whether or not a predetermined event has occurred in each of the patients. And a model generation unit. The data analysis support system according to claim 1, wherein
The storage unit further holds information indicating the cause of the occurrence of the predetermined event,
The data analysis support system according to claim 1, wherein the patient state classification unit classifies the state of the patient in which the predetermined event occurs due to the same cause to belong to the same cluster. The data analysis support system according to claim 1, wherein
The patient state classification unit includes the plurality of patients included in the medical care information in a predetermined period based on the time when the predetermined event occurs, for the patient in which the predetermined event occurs among the plurality of patients. A data analysis support system, which classifies the states of the plurality of patients into a plurality of clusters by performing a cluster analysis targeting values indicating the states of. The data analysis support system according to claim 3, wherein
The data analysis support system, wherein the patient state classification unit determines the number of clusters to be classified such that the prediction accuracy of the occurrence of the predetermined event using the prediction model is the highest. The data analysis support system according to claim 3, wherein
The patient state classification unit executes the cluster analysis multiple times while changing the number of clusters,
The prediction model generation unit
Generating the prediction model for each of the results of the cluster analysis performed multiple times;
For each of the generated prediction models, weighting factors of a plurality of factors included in the medical care information with respect to the occurrence risk of the event are acquired,
A data analysis support system characterized by outputting a prediction model having a larger weighting coefficient of a factor whose value can be changed by a treatment among a plurality of factors included in the medical care information. The data analysis support system according to claim 1, wherein
When the number of patients classified into any of the clusters is smaller than a predetermined lower limit value, the patient state classification unit reduces the number of clusters to be classified, and the patient in which the predetermined event has occurred among the plurality of patients. The plurality of patients by performing a cluster analysis on values indicating the status of the plurality of patients included in the medical care information for a predetermined period based on the time when the predetermined event occurs A data analysis support system characterized in that the state of is classified into a plurality of clusters. The data analysis support system according to claim 1, wherein
A control unit configured to predict occurrence of the predetermined event in the analysis target patient by applying the prediction model to the state of the analysis target patient extracted from the medical care information;
A data analysis support system, further comprising: an output unit that outputs a result of prediction by the control unit. A storage unit that holds history information on the states of a plurality of individuals;
An individual state classification unit that classifies the states of the plurality of individuals into a plurality of clusters based on the history information;
A prediction model is generated that predicts the occurrence of the predetermined event from the state of the classified individual based on the state of the classified individual and whether or not a predetermined event has occurred in each individual. And a model generation unit. A data analysis support method executed by a computer system having a processor and a storage unit connected to the processor, the computer system comprising:
The storage unit holds medical treatment information on the states of a plurality of patients,
The data analysis support method is
A first procedure in which the processor classifies the states of the plurality of patients into a plurality of clusters based on the medical care information;
A prediction model for predicting the occurrence of the predetermined event from the state of the classified patient based on the state of the classified patient and whether or not a predetermined event has occurred in each of the patients A second step of generating a data analysis support method. The data analysis support method according to claim 9, wherein
The storage unit further holds information indicating the cause of the occurrence of the predetermined event,
9. The data analysis support method according to claim 1, wherein in the first procedure, the processor classifies the state of the patient in which the predetermined event occurs for the same cause to belong to the same cluster. The data analysis support method according to claim 9, wherein
In the first procedure, the processor includes the medical care information included in the medical care information for a predetermined period based on a time when the predetermined event occurs for a patient in which the predetermined event occurs among the plurality of patients. A data analysis support method characterized by classifying the states of the plurality of patients into a plurality of clusters by performing a cluster analysis targeting values indicating the states of a plurality of patients. The data analysis support method according to claim 11, wherein
9. The data analysis support method according to claim 1, wherein the processor determines the number of clusters to be classified such that the prediction accuracy of the occurrence of the predetermined event using the prediction model is the highest in the first procedure. The data analysis support method according to claim 11, wherein
In the first procedure, the processor executes the cluster analysis multiple times while changing the number of clusters;
In the second procedure, the processor
Generating the prediction model for each of the results of the cluster analysis performed multiple times;
For each of the generated prediction models, weighting factors of a plurality of factors included in the medical care information with respect to the occurrence risk of the event are acquired,
A data analysis support method characterized by outputting a prediction model having a larger weighting factor of a factor whose value can be changed by a treatment among a plurality of factors included in the medical care information. The data analysis support method according to claim 9, wherein
In the first procedure, when the number of patients classified into any cluster is less than a predetermined lower limit value, the processor reduces the number of clusters to be classified, and the predetermined event of the plurality of patients is By performing cluster analysis on a patient who has occurred for a predetermined period based on the time when the predetermined event occurred, the cluster analysis is performed on values indicating the status of the plurality of patients included in the medical care information. A data analysis support method comprising classifying a plurality of patient states into a plurality of clusters. The data analysis support method according to claim 9, wherein
A third procedure in which the processor predicts the occurrence of the predetermined event in the patient to be analyzed by applying the prediction model to the condition of the patient to be analyzed extracted from the medical care information;
And D. a fourth procedure in which the processor outputs the result of the prediction.

Images