Selection support device, selection support method, and selection support program

Abstract

This selection assistance device comprises an acquisition unit, an extraction unit, an estimation unit, a prediction unit, and an output unit. The acquisition unit acquires selection criteria, which are criteria related to selection of patients to be included in a clinical trial. The extraction unit extracts patients fulfilling the selection criteria on the basis of data related to the clinical trial. The estimation unit estimates the condition of each of the extracted patients at a predetermined time point on the basis of the data related to the clinical trial. The prediction unit predicts the number of patients fulfilling the selection criteria at the predetermined time point on the basis of the estimated condition of each patient. The output unit outputs the predicted number of patients. For example, the selection assistance device can be used for assistance in decision making and the like in selection of patients to be included in a clinical trial.

Description

Selection support device, selection support method, and recording medium

[0001] The present disclosure relates to a selection support device and the like.

[0002] In clinical trials of pharmaceuticals, a person in charge at a company entrusted with a clinical trial by a pharmaceutical company will, for example, inquire with hospitals to determine the number of patients who may be suitable for the trial. The person in charge then creates a clinical trial plan, for example, based on the determined number of patients. Furthermore, the patients who will actually be included in the clinical trial are determined by carefully reviewing the details of medical records, etc., as the clinical trial plan is being developed. At this stage, if the number of patients determined in the planning stage is significantly lower than the actual number of patients, this can cause problems in the implementation of the clinical trial. Therefore, for example, at the planning stage, it is desirable to be able to easily and accurately determine the number of patients who may be suitable for the clinical trial.

[0003] The clinical trial eligible patient selection device of Patent Document 1 has a database that records medications prescribed to patients, and compares the database with clinical trial information that indicates the criteria for the clinical trial.The clinical trial eligible patient selection device of Patent Document 1 then generates a database of patients eligible for the clinical trial based on the comparison results.

[0004] JP 2016-24663 A

[0005] With the technology described in Patent Document 1, it may be difficult to predict the number of patients who will be suitable for a clinical trial.

[0006] In order to solve the above-mentioned problems, the present disclosure aims to provide a selection support device etc. that can accurately grasp the number of patients suitable for clinical trials.

[0007] In order to solve the above problems, the selection support device disclosed herein comprises an acquisition means for acquiring selection conditions, which are conditions for selecting patients for clinical trial subjects, an extraction means for extracting patients who meet the selection conditions based on data related to treatment, an estimation means for estimating the condition of each extracted patient at a specified time point based on the data related to treatment, a prediction means for predicting the number of patients who meet the selection conditions at a specified time point based on the estimated condition of each patient, and an output means for outputting the predicted number of patients.

[0008] The selection support method disclosed herein acquires selection conditions that are conditions for selecting patients for clinical trials, extracts patients who meet the selection conditions based on data related to treatment, estimates the condition of each extracted patient at a specified time point based on the data related to treatment, predicts the number of patients who meet the selection conditions at the specified time point based on the estimated condition of each patient, and outputs the predicted number of patients.

[0009] The recording medium of the present disclosure non-temporarily records a selection support program that causes a computer to execute the following processes: a process of acquiring selection conditions, which are conditions for selecting patients for clinical trial subjects; a process of extracting patients who meet the selection conditions based on data related to treatment; a process of estimating the condition of each extracted patient at a specified time point based on the data related to treatment; a process of predicting the number of patients who meet the selection conditions at a specified time point based on the estimated condition of each patient; and a process of outputting the predicted number of patients.

[0010] According to the present disclosure, the number of patients suitable for a clinical trial can be accurately determined.

[0011] FIG. 1 is a diagram illustrating an example of the configuration of a clinical trial support system according to the present disclosure. FIG. 2 is a diagram schematically illustrating an example of a data flow according to the present disclosure. FIG. 3 is a diagram illustrating an example of the configuration of a selection support device according to the present disclosure. FIG. 4 is a diagram illustrating an example of a display screen according to the present disclosure. FIG. 5 is a diagram illustrating an example of a display screen according to the present disclosure. FIG. 6 is a diagram illustrating an example of a display screen according to the present disclosure. FIG. 7 is a diagram illustrating an example of an operation flow of a selection support device according to the present disclosure. FIG. 8 is a diagram illustrating an example of an operation flow of a selection support device according to the present disclosure. FIG. 9 is a diagram illustrating an example of the hardware configuration of a selection support device according to the present disclosure.

[0012] Embodiments of the present disclosure will be described in detail with reference to the drawings. FIG. 1 is a diagram illustrating an example of the configuration of a clinical trial support system. The clinical trial support system includes a selection support device 10, a terminal device 20, a requester terminal device 30, and a data management device 40. The selection support device 10 is connected to the terminal device 20, for example, via a network. The selection support device 10 is connected to the data management device 40, for example, via a network. The terminal device 20 is connected to the requester terminal device 30, for example, via a network. The selection support device 10 may be connected to the requester terminal device 30 via a network. Furthermore, there may be multiple terminal devices 20, multiple requester terminal devices 30, and multiple data management devices 40. The number of terminal devices 20, multiple requester terminal devices 30, and multiple data management devices 40 may be set as appropriate.

[0013] A clinical trial support system is, for example, a system that supports the selection of patients to be included in a clinical trial. For example, the clinical trial support system supports the selection of patients to be included in a clinical trial by predicting the number of patients who will be eligible for the clinical trial at a given time. Patients who are eligible for the clinical trial are, for example, patients who meet the selection criteria for patients to be included in the clinical trial.

[0014] The predetermined time point is, for example, the start time of the clinical trial. For example, the start time of the clinical trial is the time when the start of the clinical trial is scheduled. In other words, the predetermined time point is a time point in the future than the time when the process of predicting the number of patients suitable for the clinical trial is performed. The predetermined time point may also be a time point during the implementation period of the clinical trial. For example, a time point during the implementation period of the clinical trial is a time point during the period when the clinical trial is scheduled to be performed. The predetermined time point is not limited to the above. For example, the condition of a patient may change between when the clinical trial plan is created and when the clinical trial is performed. Therefore, patients who were targeted for the clinical trial when the clinical trial plan was created may not meet the selection criteria when the clinical trial is performed. In such a case, it may be impossible to secure the number of patients required for the clinical trial, and data necessary for verifying the effectiveness of a new drug may not be collected. Therefore, it may be necessary to accurately determine the number of patients who meet the selection criteria when the clinical trial is performed. A clinical trial support system, for example, estimates the condition of patients at the time of the clinical trial and predicts the number of patients who meet the selection criteria when the clinical trial is performed based on the estimated results. Therefore, the clinical trial support system increases the likelihood that the clinical trial will be conducted according to plan.

[0015] The number of patients eligible for a clinical trial at a given time may be the number of patients eligible for a clinical trial after some of the selection conditions have been changed. Changing some of the selection conditions means, for example, relaxing some of the conditions among the multiple items included in the selection conditions. Relaxing a condition means, for example, broadening the range of patients indicated by the condition. For example, if the condition to be changed is age, relaxing the condition means changing the condition from 45 to 55 years old to 45 to 60 years old. Changing some of the selection conditions may also mean tightening some of the conditions among the multiple items included in the selection conditions. Changing some of the selection conditions may include changing the condition of one item in multiple stages. Tightening a condition means, for example, narrowing the range of patients indicated by the condition.

[0016] FIG. 2 is a diagram schematically illustrating an example of data flow in a clinical trial support system. In the example of FIG. 2, the selection support device 10 that extracts information about patients suitable for the clinical trial can be accessed by, for example, a person in charge at a medical institution. The person in charge at the medical institution can be, for example, one or both of a person in charge of the medical institution and a person in charge of an institution that has been entrusted with work by the medical institution. The person in charge at the medical institution is not limited to the above. In the example of FIG. 2, the selection support device 10 that extracts information about patients suitable for the clinical trial is owned, for example, by the medical institution. Being owned by the medical institution means, for example, that the person in charge at the medical institution can access the medical institution. Being owned by the medical institution can also include operating on an information system operated by an entity other than the medical institution.

[0017] In the example of Figure 2, for example, a person in charge at an institution entrusted with a clinical trial by a pharmaceutical company conducting a clinical trial of a new drug operates the requester terminal device 30 to send selection conditions to a person in charge at the medical institution, thereby inquiring about information on patients who meet the selection conditions. The institution entrusted with the clinical trial by the pharmaceutical company is, for example, a CRO (Contract Research Organization). The person in charge at the institution entrusted with the clinical trial by the pharmaceutical company is, for example, a CRA (Clinical Research Associate). The CRA may be an employee of the pharmaceutical company.

[0018] For example, the CRA inquires of a medical institution's personnel about information regarding patients who meet the selection criteria for patients to be included in the clinical trial. For example, the CRA inquires about information regarding patients who meet the selection criteria for patients to be included in the clinical trial at the stage of creating a clinical trial protocol. The clinical trial protocol is used, for example, to explain the contents of the clinical trial to and negotiate with the medical institution, and to submit notifications to related organizations. The uses of the clinical trial protocol are not limited to the above. Furthermore, the timing for inquiring about information regarding patients who meet the selection criteria is not limited to the stage of creating the clinical trial protocol.

[0019] The CRA is the person in charge at the medical institution, for example, the person in charge at the institution to which the hospital entrusts the clinical trial. The institution to which the hospital entrusts the clinical trial is, for example, an SMO (Site Management Organization). Also, the person in charge at the institution to which the hospital entrusts the clinical trial is, for example, a CRC (Clinical Research Coordinator).

[0020] The information about the patients who are the subject of the clinical trial is, for example, the number of patients who are the subject of the clinical trial. The information about the patients who are the subject of the clinical trial may be information about each patient who is the subject of the clinical trial. The information about the patients who are the subject of the clinical trial includes, for example, information about the attributes of the patients and the condition of the patients. The attributes of the patients and the condition of the patients are stored, for example, in the data management device 40.

[0021] A patient attribute is, for example, information indicating the characteristics of a patient that is related to the patient's condition and does not change with treatment, such as information on one or more of the patient's gender, age, family medical history, nationality, and race.

[0022] Information about the patient's condition is, for example, information indicating the physical condition of each patient. For example, information about the patient's condition is information indicating the patient's characteristics that may change over time since the start of treatment and due to treatment. Information about the patient's condition may include information about history. Information about the patient's condition is, for example, information on one or more items of disease information, complication information, biomarkers, disease status, guideline score, medical history, effects, and test results. Medical history is, for example, a history of treatment, drug administration, and surgery. Medical history is not limited to the above. Effects are, for example, effects of treatment, drug administration, and surgery. Effects are not limited to the above. Test results are, for example, the results of biopsies, imaging diagnostic tests, and genomic tests. Test results are not limited to the above.

[0023] The information regarding the patient's condition may be time-series data indicating the patient's condition. For example, if the information regarding the patient's condition is test results using biomarkers, the information regarding the patient's condition is time-series data of test results using the biomarkers conducted at different times. Furthermore, for example, if the information regarding the patient's condition is information whose changes can be estimated based on the patient's age, the information regarding the patient's condition does not have to be time-series data.

[0024] The CRC, for example, operates the terminal device 20 to access the selection support device 10. The selection support device 10 is a device that, for example, references data stored in the data management device 40 and extracts information about patients who are suitable for the clinical trial. The CRC then extracts information about patients who meet the selection criteria sent from the CRA. The CRC then returns the extracted patient information to the CRA. The CRA, for example, references the patient information included in the response from the CRC and creates a clinical trial protocol.

[0025] 2 shows an example in which the selection support device 10 is owned by a medical institution, but the selection support device 10 may be operated by a third party. A third party refers to an entity other than a pharmaceutical company, an institution commissioned by a pharmaceutical company, a medical institution, or an institution commissioned by a medical institution. In this case, the entity operating the selection support device 10 receives data on the patient's condition from, for example, the medical institution. The selection support device 10 extracts and outputs information on patients suitable for a clinical trial based on a request from, for example, a terminal device 20 used by a medical institution's staff member or a requester terminal device 30 used by a pharmaceutical company's staff member.

[0026] The selection support device 10 may also be owned by a pharmaceutical company or an institution commissioned by a pharmaceutical company. In this case, the selection support device 10 extracts and outputs information about patients suitable for the clinical trial by, for example, referencing pseudonymized data related to treatment stored in the data management device 40. Examples of the configuration of the clinical trial support system are not limited to those described above. The configuration of the clinical trial support system may be set as appropriate.

[0027] Here, a specific example of the configuration of the selection support device 10 will be described. Fig. 3 is a diagram showing an example of the configuration of the selection support device 10. The selection support device 10 basically includes an acquisition unit 11, an extraction unit 12, an estimation unit 13, a prediction unit 15, and an output unit 16. The selection support device 10 also includes, for example, a compatibility calculation unit 14 and a storage unit 17.

[0028] The acquisition unit 11 acquires selection conditions, which are conditions related to the selection of patients to be included in the clinical trial. The selection conditions are, for example, conditions that indicate the range of patients to be extracted when extracting patients from patients for whom data related to treatment is stored. The selection conditions are, for example, conditions for including patients in the clinical trial. The conditions for including patients in the clinical trial are conditions that indicate patients who will be included in the clinical trial if the conditions are met. The selection conditions may also be exclusion conditions, which are conditions for patients to be excluded from the clinical trial. The exclusion conditions are conditions that indicate patients who will not be included in the clinical trial if the conditions are met. Furthermore, the selection conditions may be both conditions for including patients in the clinical trial and exclusion conditions.

[0029] The selection conditions are, for example, conditions set based on multiple items of information included in the information on the patient's treatment. The information on the patient's treatment is, for example, information on the patient's attributes and the patient's condition. The selection conditions may include conditions set based on multiple items included in the patient's attributes. The selection conditions may also include conditions set based on multiple items included in the information on the patient's condition.

[0030] The extraction unit 12 extracts patients who meet the selection conditions based on, for example, data related to treatment. The extraction unit 12 may extract patients who meet the selection conditions and patients who meet conditions obtained by partially modifying the selection conditions based on the data related to treatment. The data related to treatment is, for example, a record of medical procedures performed on the patient. The data related to treatment may also include patient attributes. The extraction unit 12 extracts patients who meet the selection conditions based on, for example, data related to treatment stored as an electronic medical record in the data management device 40. The extraction unit 12 also extracts patients who meet conditions obtained by partially modifying the selection conditions based on, for example, data related to treatment stored as an electronic medical record in the data management device 40. The data related to treatment is not limited to data stored outside of an electronic medical record.

[0031] For example, the extraction unit 12 refers to the treatment-related data of each patient stored in the data management device 40 and extracts patients who meet the selection conditions. Furthermore, the extraction unit 12 extracts, as patients who meet the selection conditions, patients whose treatment-related data matches indices indicating the range of patients included in each of the conditions included in the items specified in the selection conditions. For example, the extraction unit 12 extracts patients whose treatment-related data falls within the range of indices specified in each of the items specified in the selection conditions.

[0032] The extraction unit 12 may extract patients based on the suitability of the treatment-related data for the selection conditions. For example, the extraction unit 12 outputs patients whose suitability for the selection conditions is equal to or exceeds a standard as patients who meet the selection conditions. The suitability for each patient is calculated, for example, by the suitability calculation unit 14. The suitability for each patient may be calculated based on the difference between the median value of the selection conditions and the treatment-related data for each patient, and the weight of each selection condition item. The weight of each selection condition item is set, for example, based on the magnitude of the influence that each selection condition item may have on the results of the clinical trial. For example, if the efficacy of the drug being tested is significantly affected by the drug administration history, the weight of each selection condition item is set so that the drug administration history is weighted higher than the other items.

[0033] The extraction unit 12 may use an extraction model to extract patients who meet the selection conditions. For example, the extraction unit 12 uses an extraction model generated by machine learning based on data related to treatment to extract patients who meet the selection conditions and patients who meet conditions obtained by changing some of the items included in the selection conditions. That is, the extraction model is, for example, a learning model that uses the selection conditions as input and extracts patients who meet the selection conditions. For example, the extraction model extracts patients who meet the selection conditions by extracting patients whose data related to treatment is similar to the selection conditions. Furthermore, the extraction model is, for example, generated by deep learning using a neural network.

[0034] The extraction model may be generated as an optimization model using mathematical optimization. When the extraction model is an optimization model, for example, a constraint condition is set such that the number of extracted patients is equal to or greater than a predetermined number. In this case, the extraction model is generated to extract patients that meet some of the selection conditions, which are modified so that the number of extracted patients is equal to or greater than the predetermined number and the average value of the degree of conformance is maximized. The predetermined number is, for example, the number of patients required for the clinical trial. Furthermore, an item that cannot be changed among the selection conditions may be further specified as a constraint condition. The algorithm for generating the extraction model is not limited to the above. The extraction model may be generated, for example, in an information processing device external to the selection support device 10. The extraction model may also be generated in a learning unit (not shown) within the selection support device 10.

[0035] The extraction unit 12 may extract patients who meet conditions obtained by changing some of the selection conditions based on the results of estimating the state of each extracted patient at a predetermined time point. The estimation of the patient's state at a predetermined time point is performed, for example, by the estimation unit 13. The predetermined time point is, for example, the start of the clinical trial. The predetermined time point may be a time point during the clinical trial period. There may also be multiple predetermined time points. For example, the predetermined time point may be multiple time points during the clinical trial period. The predetermined time point may be the start of the clinical trial and a time point during the clinical trial period. There may also be multiple predetermined time points at regular intervals from the time of patient extraction. When there are multiple predetermined time points, the intervals between the time points do not have to be equal. The predetermined time points are not limited to those described above.

[0036] The extraction unit 12 may also extract patients who meet the selection conditions in multiple stages. For example, the extraction unit 12 extracts patients who meet the selection conditions in two stages. For example, the extraction unit 12 extracts patients based on a first condition, which is a required condition among the selection conditions. A required condition is, for example, a condition that cannot be changed for the purpose of the clinical trial. Then, the extraction unit 12 extracts patients who meet the selection conditions from among the patients extracted based on the first condition based on a second condition, which is other than the first condition among the selection conditions. Furthermore, the extraction unit 12 may extract patients who meet a condition changed from the second condition based on data related to treatment from among the patients extracted based on the first condition.

[0037] The extraction unit 12 may extract patients who meet the second condition from among the patients extracted based on the first condition, based on the result of estimating the condition of each of the extracted patients at a predetermined time point. Furthermore, the extraction unit 12 may extract patients who meet a condition that is relaxed from the second condition from among the patients extracted based on the first condition, based on the result of estimating the condition of each of the extracted patients at a predetermined time point. In other words, the extraction unit 12 extracts patients who meet a condition obtained by changing part of the second condition from the conditions included in the selection conditions, for example.

[0038] The extraction unit 12 may change some of the selection conditions until the total number of patients whose suitability meets the criteria is equal to or greater than the number required for the clinical trial, and extract patients who meet the changed conditions. The extraction unit 12 may change some of the selection conditions, for example, based on the priority of selecting conditions to be changed from among the conditions included in the selection conditions. The extraction unit 12, for example, changes some of the selection conditions to extract patients who meet the changed conditions. In this case, if the number of extracted patients is equal to or greater than the required number of patients, the result becomes the extraction result by the extraction unit 12. Furthermore, if the number of patients is insufficient for the number required for the clinical trial, the extraction unit 12 further changes some of the selection conditions and re-extracts patients who meet the changed conditions. The extraction unit 12 appropriately changes the conditions and repeats the extraction until the number of patients is equal to or greater than the number required for the clinical trial. Furthermore, when extracting patients so that the total number of patients whose suitability meets the criteria is equal to or greater than the number required for the clinical trial, the extraction unit 12 extracts patients who meet the changed conditions for each of the multiple items other than the items that cannot be changed.

[0039] The treatment data may be pseudonymized data. That is, the extraction unit 12 may extract patients who meet the selection conditions from the pseudonymized treatment data. The pseudonymized treatment data is, for example, data from which information identifying each patient has been deleted, but from which each patient can be identified by matching with other information. The treatment data may also be anonymized data. The anonymized treatment data is, for example, data from which information identifying each patient has been deleted, and from which each patient cannot be identified even when matched with other information.

[0040] The estimation unit 13 estimates the condition of each extracted patient at a predetermined time point based on data related to treatment. The estimation unit 13 estimates the condition of each patient extracted by the extraction unit 12 at a predetermined time point based on data related to treatment. The estimation unit 13 may also estimate the condition of each patient set as an estimation target at a predetermined time point. The setting of patients to be estimated is performed, for example, in a device external to the selection support device 10. The setting of patients to be estimated may also be performed at a timing different from the timing of patient extraction by the extraction unit 12. The setting of patients to be estimated may be performed, for example, by a user selecting patients to be estimated at the start of processing related to predicting the number of patients who meet the selection conditions at a predetermined time point. The estimation unit 13 estimates the condition of each patient at the time when a clinical trial is started, for example.

[0041] The predetermined time point is set, for example, based on the time point when the clinical trial is scheduled to be conducted. For example, a patient who currently meets the selection conditions may not meet the selection conditions when the clinical trial is conducted. For this reason, for example, by estimating the patient's condition at the start of the clinical trial and extracting patients who meet the selection conditions based on the estimation results, it is possible to increase the likelihood that the extracted patients will meet the selection conditions when the clinical trial is conducted. Furthermore, the predetermined time point may be set based on the rate of progression of the disease targeted by the new drug. The predetermined time point may be set based on the type of new drug. The predetermined time point may be set based on the season in which the disease targeted by the new drug is likely to worsen. The criteria for setting the predetermined time point are not limited to the above.

[0042] The estimation unit 13 may estimate the condition of each patient at multiple time points. For example, the estimation unit 13 estimates the condition of each patient one month, three months, and five months after the time point of estimation. Furthermore, for example, the estimation unit 13 may estimate the condition of each patient one month, three months, six months, and 12 months after the time point of estimation. The multiple time points are not limited to the above.

[0043] The predetermined time point is set, for example, by a user of the selection support device 10. The user of the selection support device 10 is, for example, a person who has access rights to the selection support device 10. The user of the selection support device 10 is, for example, a CRC. The user of the selection support device 10 may also be a CRA. The user of the selection support device 10 may also be someone other than a CRC or a CRA. The user of the selection support device 10 sets the predetermined time point, for example, based on the time point when the clinical trial is scheduled to start.

[0044] The estimation unit 13 estimates the condition of patients who are candidates for clinical trial subjects, for example, using an estimation model generated by machine learning that estimates the patient's condition at a specified time point from the patient's attributes and data related to the patient's treatment.

[0045] The estimation model estimates the patient's condition at a predetermined time point using, for example, time-series data related to the patient's attributes and treatment as input. The estimation model is generated, for example, by learning the relationship between the time-series data related to the patient's attributes and treatment and the patient's future condition. The estimation model is generated, for example, by deep learning using a neural network. The learning algorithm for generating the estimation model is not limited to the above. The estimation model is generated, for example, in an information processing device external to the selection support device 10. The estimation model may also be generated in a learning unit (not shown) within the selection support device 10.

[0046] The estimation unit 13 may estimate the patient's condition at a predetermined time point from the data related to the patient's treatment by performing regression analysis. For example, the estimation unit 13 estimates the patient's condition at a predetermined time point from the data related to the patient's treatment by performing multiple regression analysis.

[0047] The suitability calculation unit 14 calculates, for example, the suitability of the extracted patient to the selection conditions. The suitability to the selection conditions is, for example, an index indicating the degree to which information about the patient's treatment matches the selection conditions. That is, the suitability to the selection conditions is, for example, an index indicating the degree to which the patient is suitable for the clinical trial. When the selection conditions are exclusion conditions, the suitability to the selection conditions may be, for example, an index indicating the degree to which the patient is not suitable for the clinical trial. Furthermore, the suitability calculation unit 14 may calculate the suitability of each patient extracted under the changed conditions whose suitability does not satisfy the criteria, under conditions in which some of the selection conditions are made stricter. When the suitability satisfies the criteria by making some of the selection conditions stricter, for example, the output unit 16 outputs the conditions in which some of the selection conditions are made stricter as the conditions necessary for the suitability to satisfy the criteria.

[0048] The suitability calculation unit 14 may also calculate the suitability of each patient to the selection conditions based on the patient's condition at a predetermined time estimated by the estimation unit 13. In this case, the suitability to the selection conditions is, for example, an index indicating the degree to which the patient's estimated condition at a predetermined time matches the selection conditions. The suitability calculation unit 14 may also calculate the suitability to the selection conditions for each patient extracted by the extraction unit 12 based on the patient's condition at a predetermined time as a patient who matches a condition obtained by partially changing the selection conditions. The suitability calculation unit 14 calculates the suitability of the extracted patient to the selection conditions using, for example, a suitability calculation model. The suitability calculation model is, for example, a learning model that calculates the suitability using the patient's attributes and patient condition as selection conditions as input.

[0049] The compatibility calculation model converts, for example, the selection conditions, the patient attributes and the patient conditions of the selection conditions, and each of the selection conditions into a feature vector.The compatibility calculation model then calculates the compatibility by, for example, calculating the Euclidean distance between the feature vectors.The distance between the feature vectors is not limited to the Euclidean distance.In addition, when the information representing the patient attributes and the patient conditions is text information, the compatibility calculation model converts the information representing the patient attributes and the patient conditions into an embedding vector.

[0050] The fitness calculation model is generated, for example, by deep learning using a neural network. The learning algorithm for generating the fitness calculation model is not limited to the above. The fitness calculation model is generated, for example, in an information processing device external to the selection support device 10. The fitness calculation model may also be generated in a learning unit (not shown) within the selection support device 10.

[0051] Furthermore, the fitness calculation model may be selected from a plurality of fitness calculation models. In this case, the weights of the feature quantities used in each fitness calculation model may be changeable by, for example, the user. For example, the weights of the feature quantities are set so that the weights corresponding to items that are given more importance in patient selection are increased. Furthermore, for example, the weights of the feature quantities corresponding to items that are given less importance than other items in patient selection are set so that the weights are decreased.

[0052] The compatibility calculation unit 14 may calculate a score based on a criterion set for each item of the selection conditions. The score is, for example, an index showing the degree of compatibility of data related to the patient's condition with each item included in the selection conditions. The criterion for calculating the score is set, for example, based on the patient's attribute and patient's condition data for each item of the selection conditions and the deviation from the central value of the selection conditions. The criterion for calculating the score is set, for example, based on the importance of each item of the selection conditions and the allowable range of deviation. The criterion for calculating the score is set, for example, using a table showing the relationship between the selection criteria, the patient's attribute and patient's condition data, and the score. The criterion for calculating the score may be set using a function.

[0053] The compatibility calculation unit 14 calculates the compatibility by, for example, multiplying the scores of the respective selection condition items. The compatibility calculation unit 14 may calculate the compatibility by adding the scores of the respective selection condition items. The compatibility calculation unit 14 may calculate the compatibility by, for example, subtracting the score of each item from the maximum score. Furthermore, the compatibility calculation unit 14 may calculate the compatibility by weighting the score of each selection condition item based on the importance of the item.

[0054] The prediction unit 15 predicts the number of patients who meet the selection conditions at a predetermined time point based on the condition of each patient estimated by the estimation unit 13. The prediction unit 15, for example, counts the number of patients who meet the selection conditions in their condition at a predetermined time point. Then, the prediction unit 15, for example, sets the result of counting the number of patients as the predicted result of the number of patients. Furthermore, when the estimation unit 13 estimates the condition of each patient at multiple time points, the prediction unit 15, for example, predicts the number of patients who meet the selection conditions at each of the multiple time points. By predicting the number of patients who meet the selection conditions at each of the multiple time points, it is possible to grasp, for example, a period when there are many patients who meet the clinical trial based on the prediction result.

[0055] The prediction result may also be the number of patients who meet the selection conditions with some of the selection conditions changed. For example, the extraction unit 12 extracts patients who meet the selection conditions and patients who meet the selection conditions with some of the selection conditions changed, based on data related to treatment. The estimation unit 13 estimates the condition of each of the patients extracted by the extraction unit 12 at a predetermined time point, for example, based on the data related to treatment. Then, the prediction unit 15 predicts the number of patients who meet the selection conditions at a predetermined time point, based on the condition of each patient estimated by the estimation unit 13. For example, the extraction unit 12 extracts patients who meet the selection conditions based on data related to treatment. The estimation unit 13 estimates the condition of each of the patients extracted by the extraction unit 12 at a predetermined time point, for example, based on the data related to treatment. The extraction unit 12 extracts patients who meet the selection conditions and patients who meet the selection conditions with some of the selection conditions changed, based on the condition of each patient estimated by the estimation unit 13 at a predetermined time point. The prediction unit 15 then predicts the number of patients who meet the selection conditions at a predetermined time point from the result of the extraction unit 12 extracting patients based on the condition of each patient at the predetermined time point.

[0056] The prediction unit 15 may predict the number of patients meeting the selection conditions at a predetermined time point, as well as a fluctuation range from the predicted value of the number of patients. The prediction unit 15 predicts a fluctuation range from the predicted value of the number of patients, for example, based on the reliability of the estimation result of the patient's condition using the estimation model. The prediction unit 15 predicts a fluctuation range of the patient's condition, for example, based on the reliability of the estimation result of the patient's condition using the estimation model. The prediction unit 15 also calculates the number of patients meeting the selection conditions when the condition of each patient changes by the fluctuation range. The prediction unit 15 then determines the fluctuation range as the range between the maximum and minimum values ​​of the number of patients meeting the selection conditions when the condition of each patient changes by the fluctuation range. The relationship between the fluctuation range from the predicted value of the number of patients may be set using a function using the predicted value and the reliability as explanatory variables. The relationship between the fluctuation range from the predicted value of the number of patients may be set as a table showing the relationship between the predicted value, the reliability, and the predicted value of the fluctuation range. The relationship between the predicted value and the fluctuation range is set, for example, so that the more reliable the predicted value is, the smaller the fluctuation range becomes. The relationship between the fluctuation range and the predicted value is set so that, for example, the lower the probability of the predicted value, the larger the fluctuation range.

[0057] The output unit 16 outputs the number of patients predicted by the prediction unit 15. The output unit 16 outputs, for example, the number of patients who meet the selection conditions at a predetermined time point. The output unit 16 may output the number of patients predicted by the prediction unit 15 in association with a fluctuation range of the prediction. The output unit 16 may output the number of patients predicted by the prediction unit 15 in association with the likelihood of the prediction. The likelihood of the prediction may be, for example, the accuracy of the estimation result of the patient's condition using the estimation model. Furthermore, the output unit 16 may output the number of patients who meet the selection conditions at a predetermined time point, predicted using the extraction result of patients who meet conditions obtained by changing some of the selection conditions.

[0058] When the number of patients is predicted at multiple time points, the output unit 16 may output time-series data of the number of patients predicted by the prediction unit 15. Furthermore, the output unit 16 may output the prediction result of the number of patients at each of the multiple time points in association with the prediction fluctuation range. Furthermore, the output unit 16 may output, for example, statistical data regarding the conditions of multiple patients who meet the selection conditions at a predetermined time point. The statistical data regarding the conditions of multiple patients who meet the selection conditions is statistical data for each of the patient conditions. For example, if the patient condition is age, the statistical data is the average value of the age. Furthermore, if the patient condition is the elapsed time since onset, the statistical data is the average value of the elapsed time since onset. The statistical data is not limited to the average value. The output unit 16 may output the patient condition at a predetermined time point estimated by the estimation unit 13 for each patient who meets the selection conditions.

[0059] The output unit 16 outputs, for example, information regarding the suitability of patients extracted under the changed conditions. The information regarding the suitability of patients is, for example, information indicating a breakdown of patients extracted under the changed conditions. The information regarding the suitability of patients extracted under the changed conditions is, for example, the number of patients extracted under the changed conditions. Furthermore, the information regarding the suitability of patients extracted under the changed conditions may be the suitability of each patient for the selection conditions. Furthermore, the information regarding the suitability of patients extracted under the changed conditions may be both the number of patients extracted under the changed conditions and the suitability of each patient for the selection conditions.

[0060] The output unit 16 outputs, for example, to the terminal device 20, information regarding the suitability of patients extracted under the changed conditions. The output unit 16 may output, for example, information regarding the suitability of patients extracted under the changed conditions to the client terminal device 30. The output unit 16 outputs, for example, the number of patients extracted under the changed conditions and information regarding the suitability of each patient. The output unit 16 may output the number of patients extracted under the changed conditions and statistical values ​​of the suitability of patients increased by the change. The output unit 16 outputs, for example, the number of patients extracted under the changed conditions and the average value of the suitability of patients increased by the change. The statistical value is not limited to the average value.

[0061] The output unit 16 may output the number of patients increased by changing each of the selection conditions and the cumulative number of patients. The cumulative number of patients is the number of patients obtained by sequentially adding up the number of patients increased by changing some of the selection conditions when the selection conditions are repeatedly changed. For example, if the number of extracted patients increases by seven from 40 due to a change in the conditions, the output unit 16 outputs, for example, that the number of patients increased by seven due to the change in the conditions and that the cumulative number of patients is 47.

[0062] The output unit 16 may further output a display screen for the user to input information on the screen. For example, when the user sets weights for each item included in the selection conditions, the output unit 16 may output a display screen for inputting a set value of the weight for each item included in the selection conditions. Furthermore, when the user selects items to be changed among the selection conditions, the output unit 16 may output a display screen for inputting the items to be changed. As a display screen for inputting the items to be changed, the output unit 16 may output a display screen for selecting the items to be changed from the displayed items of the selection conditions. The output unit 16 may output a display screen for inputting the degree of change for each item to be changed.

[0063] FIG. 4 is an example of a display screen displaying the predicted results of the number of patients at a given time point. The example display screen of FIG. 4 displays the selection criteria related to disease and age. The example display screen of FIG. 4 also displays the predicted values ​​of the number of patients eligible for the clinical trial in August, October, and February of the following year, and the range of the predicted values ​​as a fluctuation range. The example display screen of FIG. 4 displays the predicted results of the number of patients in August, October, and February of the following year, as well as the predicted results of the number of patients one month, three months, and six months later, and the fluctuation range of the predicted results.

[0064] FIG. 5 is an example of a display screen displaying the predicted results of the number of patients at a given time point. The example display screen of FIG. 5 displays the selection criteria related to disease and age. The example display screen of FIG. 5 also displays the predicted values ​​of the number of patients eligible for the clinical trial in August, October, and February of the following year, and the range of the predicted values ​​as a fluctuation range. The example display screen of FIG. 5 displays the predicted results of the number of patients in August, October, and February of the following year, as well as the predicted results of the number of patients one month, three months, and six months later, and the fluctuation range of the predicted results.

[0065] FIG. 6 is an example of a display screen showing the results of patient extraction when patients are extracted by changing some of the selection conditions. In the example of the display screen in FIG. 6, the number of patients who meet the selection conditions and the number of patients who meet the modified selection conditions are displayed as the number of candidates. In the example of the display screen in FIG. 6, the number of patients who meet the selection conditions is displayed as 12, and the number of patients who meet the modified selection conditions is displayed as 17. In the example of the display screen in FIG. 6, the age condition, which was set as between 45 and 55 years old, has been changed to between 40 and 60 years old. In addition, the example of the display screen in FIG. 6 also displays that the number of patients increased by 5 by changing the age condition.

[0066] The example display screen of FIG. 6 may also be a display screen showing a predicted result of the number of patients at a predetermined time point. The predetermined time point is, for example, the start time of a clinical trial. For example, the extraction unit 12 extracts patients who meet the selection conditions and patients who meet a modified version of the selection conditions based on data related to treatment. The estimation unit 13 estimates the condition of each patient extracted by the extraction unit 12 at a predetermined time point based on data related to treatment. The prediction unit 15 predicts the number of patients who meet the selection conditions at a predetermined time point based on the condition of each patient estimated by the estimation unit 13. The output unit 16 then outputs the predicted result as shown in the example display screen of FIG. 6. The extraction unit 12 may also extract patients who meet the selection conditions and patients who meet a modified version of the selection conditions based on the condition of each patient estimated by the estimation unit 13. In this case, the prediction unit 15 predicts the number of patients who meet the selection conditions at a predetermined time point using the extraction result based on the condition of each patient estimated by the estimation unit 13. The output unit 16 then outputs the predicted result as shown in the example display screen of FIG. 6.

[0067] FIG. 7 is an example of a display screen showing the results of patient extraction when patients are extracted by changing some of the selection conditions. In the example of the display screen of FIG. 7 , the number of patients who meet the selection conditions and the number of patients who meet the modified selection conditions are displayed as the number of candidates. In the example of the display screen of FIG. 7 , the number of patients who meet the modified selection conditions and the average fitness scores of the patients are displayed. In the example of the display screen of FIG. 7 , the number of patients who meet the modified selection conditions is displayed as 12 and 17. In the example of the display screen of FIG. 7 , the age condition, which was set as 45 to 55 years old, has been changed to 40 to 60 years old. In addition, the example of the display screen of FIG. 7 shows that the number of patients increased by 5 by changing the age condition and that the average fitness score of the 5 patients is 85. The example of the display screen of FIG. 7 may also be a display screen showing a predicted number of patients at a given time point.

[0068] FIG. 8 shows an example of a display screen for setting the priority of selection conditions. The priority of a selection condition is the priority for selecting an item to be changed when changing the conditions for some of the selection conditions. In the example of the display screen of FIG. 8, the priority of the selection condition related to disease is set to unchangeable. Also, in the example of the display screen of FIG. 8, the selection condition related to age is set to "8." The priority is set, for example, so that the larger the value, the higher the priority for change. Also, in the example of the display screen of FIG. 8, an "Update" button is displayed. When a user inputs a priority value on the display screen and then presses the "Update" button, the priority is updated based on the input value. The priority is input, for example, by the user operating the terminal device 20. Then, the acquisition unit 11 acquires the priority input by the user from, for example, the terminal device 20. The extraction unit 12, for example, changes some of the selection conditions based on the acquired priority until the total number of patients whose suitability criteria is met is equal to or greater than the number required for the clinical trial. Then, for example, each time the conditions are changed, the extraction unit 12 extracts patients who meet the changed conditions.

[0069] The memory unit 17 stores, for example, data related to the prediction of the number of patients at a predetermined time point. The memory unit 17 stores, for example, the patient extraction results under selection conditions and conditions that partially change the selection conditions. The memory unit 17 also stores, for example, the prediction results of the number of patients at a predetermined time point. The memory unit 17 may also store, for example, the weight setting values ​​for each item of the selection conditions. The memory unit 17 stores, for example, extraction models and estimation models. The memory unit 17 also stores a goodness-of-fit calculation model. The extraction model, estimation model, and goodness-of-fit calculation model may be stored in a storage means external to the selection support device 10.

[0070] The terminal device 20 is, for example, a terminal device that accesses the selection support device 10 and is used for processing to extract candidate patients suitable for the clinical trial. The terminal device 20 is, for example, a terminal device used by a person in charge of a medical institution or a person in charge of an institution entrusted with work related to the clinical trial by the medical institution. The person in charge of an institution entrusted with work related to the clinical trial by the medical institution is, for example, a CRC.

[0071] The terminal device 20 outputs the selection conditions to, for example, the acquisition unit 11 of the selection support device 10. Then, the terminal device 20 acquires information on the suitability of patients extracted under the changed conditions from the output unit 16 of the selection support device 10. Then, the terminal device 20 outputs information on the suitability of patients extracted under the changed conditions to a display device (not shown) connected to the terminal device 20.

[0072] The terminal device 20 also acquires, for example, selection conditions related to the selection of patients to be included in the clinical trial from the requester terminal device 30. The terminal device 20 may also acquire, for example, the number of patients required as patients to be included in the clinical trial and the timing of the clinical trial from the requester terminal device 30. The terminal device 20 also outputs, for example, information regarding the suitability of patients extracted under the changed conditions to the requester terminal device 30. The terminal device 20 also outputs, for example, the predicted number of patients to the requester terminal device 30.

[0073] The requester terminal device 30 is a terminal device used, for example, by a pharmaceutical company representative or a representative of an institution entrusted with clinical trial work by a pharmaceutical company. The representative of an institution entrusted with clinical trial work by a pharmaceutical company is, for example, a CRA. The requester terminal device 30 is used, for example, to request information on patients who are the subject of clinical trials from medical institutions. The requester terminal device 30 is also used, for example, to obtain information on patients who meet the selection criteria provided by medical institutions.

[0074] The requester terminal device 30 outputs, for example, to the terminal device 20, selection conditions related to the selection of patients to be included in the clinical trial. The requester terminal device 30 may also output, for example, to the terminal device 20, the number of patients required as clinical trial patients and the timing of the clinical trial. The requester terminal device 30 also acquires, for example, from the terminal device 20, information regarding the suitability of patients extracted under the changed conditions. The requester terminal device 30 then displays, for example, information regarding the suitability of patients extracted under the changed conditions on a display device (not shown) connected to the requester terminal device 30. The requester terminal device 30 also acquires, for example, a predicted number of patients from the terminal device 20. The requester terminal device 30 then displays, for example, the predicted number of patients on a display device (not shown) connected to the requester terminal device 30.

[0075] The requester terminal device 30 may be a terminal device that accesses the selection support device 10 and is used for processing to extract candidate patients suitable for the clinical trial. The requester terminal device 30 outputs the selection conditions to the acquisition unit 11 of the selection support device 10, for example. The requester terminal device 30 then acquires information regarding the suitability of patients extracted under the changed conditions from the output unit 16 of the selection support device 10. The requester terminal device 30 then outputs information regarding the suitability of patients extracted under the changed conditions to a display device (not shown).

[0076] The data management device 40 is a device for storing data related to patient treatment. The data related to patient treatment is, for example, data in an electronic medical record entered by a doctor. The data related to patient treatment may be information on one or more items of the patient's disease information, complication information, biomarkers, disease status, guideline score, medical history, effects, and test results other than those described in the electronic medical record.

[0077] A description will be given of the process of predicting the number of candidates for a clinical trial at a predetermined time in the selection support device 10. Fig. 9 shows an example of the operational flow of the process of predicting the number of candidates for a clinical trial at a predetermined time.

[0078] The acquisition unit 11 acquires selection conditions, which are conditions related to the selection of patients to be subject to the clinical trial (step S11). The acquisition unit 11 acquires the selection conditions, which are conditions related to the selection of patients to be subject to the clinical trial, from the terminal device 20, for example.

[0079] When the selection conditions are acquired, the extraction unit 12 extracts patients who meet the selection conditions based on the data related to treatment (step S12).

[0080] When the patients who meet the selection conditions are extracted, the estimation unit 13 estimates the condition of each extracted patient at a predetermined time point based on the data related to treatment (step S13).

[0081] When estimation has been completed for all extracted patients (Yes in step S14), the prediction unit 15 predicts the number of patients who meet the selection conditions at a predetermined time point based on the estimated condition of each patient (step S15). In step S14, when estimation has been completed for all extracted patients, the extraction unit 12 may extract patients who meet the selection conditions and patients who meet a modified version of the selection conditions based on the condition of each patient at a predetermined time point estimated by the estimation unit 13. When the extraction unit 12 extracts patients based on the condition of each patient at a predetermined time point, for example, in step S15, the prediction unit 15 predicts the number of patients who meet the selection conditions at a predetermined time point using the extraction result of patients who meet a modified version of the selection conditions.

[0082] When the number of patients is predicted, the output unit 16 outputs the number of patients predicted by the prediction unit 15 (step S16). Furthermore, if the extraction result of patients who meet the selection conditions with some of the selection conditions changed is used in step S15, for example, in step S15, the output unit 16 outputs the number of patients who meet the selection conditions at a predetermined time point, predicted using the extraction result of patients who meet the selection conditions with some of the selection conditions changed.

[0083] In step S14, if there are patients for whom estimation has not been completed (No in step S14), the process returns to step S13, and the estimation unit 13 estimates the condition of the patients for whom estimation has not been completed at a specified time point, for example, based on data related to treatment.

[0084] A description will be given of the process of extracting candidates for clinical trials by relaxing the selection conditions in the selection support device 10. Fig. 10 shows an example of the operational flow of the process of extracting candidates for clinical trials by relaxing the selection conditions.

[0085] The acquisition unit 11 acquires, for example, selection conditions, which are conditions related to the selection of patients to be subject to a clinical trial (step S21). The acquisition unit 11 acquires, for example, the selection conditions, which are conditions related to the selection of patients to be subject to a clinical trial, from the terminal device 20.

[0086] When the selection conditions are acquired, the extraction unit 12 extracts patients who meet the selection conditions and patients who meet conditions obtained by partially modifying the selection conditions, for example, based on data related to treatment (step S22). The extraction unit 12 may extract patients who meet the selection conditions and patients who meet conditions obtained by partially modifying the selection conditions, using the conditions of each patient at a predetermined time point estimated by the estimation unit based on the data related to treatment.

[0087] When the patient is extracted, the compatibility calculation unit 14 calculates, for example, the compatibility of the extracted patient with the selection conditions (step S13).

[0088] If the number of extracted patients is equal to or greater than the number of patients required for the clinical trial (Yes in step S24), the output unit 16 outputs, for example, information about the suitability of the patients extracted under the changed conditions (step S25). The output unit 16 outputs, for example, information about the suitability of the patients extracted under the changed conditions to the terminal device 20.

[0089] Furthermore, in step S24, if the number of extracted patients is less than the number of patients required for the clinical trial (No in step S24), the extraction unit 12, for example, changes the conditions of the selection conditions that have not been changed, and extracts patients who meet the changed conditions (step S26). For example, the extraction unit 12 changes some of the selection conditions until the total number of patients whose suitability meets the criteria is equal to or greater than the number required for the clinical trial, and extracts patients who meet the changed conditions.

[0090] If the number of extracted patients does not exceed the number of patients required for the clinical trial even after changing some of the selection conditions, the extraction unit 12 may designate the extracted patients as the finally extracted patients. For example, if the number of extracted patients does not exceed the number of patients required for the clinical trial even after changing all conditions other than those set as unchangeable, the extraction unit 12 designates the extracted patients as the finally extracted patients. Then, the output unit 16 outputs, for example, information regarding the suitability of patients extracted under the changed conditions based on the finally extracted patients. Furthermore, the extraction unit 12 may change each of the conditions included in the selection conditions in multiple stages.

[0091] When the patient is extracted in step S26, the process returns to step S23, and the compatibility calculation unit 14 calculates the compatibility of the extracted patient with the selection conditions based on the changed conditions.

[0092] The selection support device 10 extracts patients who meet the selection conditions based on data related to treatment. The selection support device 10 estimates the condition of each extracted patient at a predetermined time point based on the data related to treatment. The selection support device 10 predicts the number of patients who meet the selection conditions at a predetermined time point based on the estimated condition of each patient. The selection support device 10 then outputs the predicted number of patients. In this way, by predicting the number of patients who meet the selection conditions based on the results of estimating the condition of each patient at a predetermined time point, it is possible to accurately determine the number of patients who are suitable for a clinical trial at a predetermined time point. For example, by predicting the number of patients who are suitable for a clinical trial at the start of the clinical trial based on the estimated results of the patient's condition, it is possible to prevent a situation in which there are no suitable patients when the clinical trial begins. Therefore, for example, by predicting the number of patients who are suitable for a clinical trial at the start of the clinical trial based on the estimated results of the patient's condition, it is possible to increase the likelihood that the clinical trial will be conducted according to the actual plan.

[0093] Furthermore, for example, by outputting the predicted number of patients in association with the range of fluctuation in the prediction, a person using the prediction result can use the prediction result while recognizing the range within which the number of patients may fluctuate.

[0094] Furthermore, for example, by predicting the number of patients who meet the selection criteria at each of multiple time points, it is possible to determine the period when there will be many patients who are suitable for the clinical trial based on the prediction results. Therefore, for example, it is possible to select the most suitable period for conducting the clinical trial based on the prediction results and create a clinical trial plan.

[0095] The selection support device 10, for example, extracts patients who meet the selection conditions and patients who meet conditions obtained by changing some of the items included in the selection conditions based on data related to treatment. The selection support device 10, for example, calculates the suitability of the extracted patients for the selection conditions. Then, the selection support device 10, for example, outputs information regarding the suitability of the patients extracted under the changed conditions. In this way, by outputting information regarding the suitability of patients when some of the selection conditions are changed, it becomes possible to determine the suitability of the increasing number of candidate clinical trial subjects for the clinical trial when the conditions are changed, for example. This makes it easy to select candidate clinical trial subjects.

[0096] For example, by outputting the number of patients who are candidates for clinical trial subjects that has increased as a result of changing the conditions and the degree of conformity of the increased patients to the selection conditions, it is possible to select patients who are candidates for clinical trial subjects while taking into consideration the impact of changing the conditions on the clinical trial. Also, for example, by extracting patients based on the priority of items to be changed among the items included in the selection conditions, so that the total number of extracted patients is equal to or greater than the number required for the clinical trial, it is possible to easily select the number of patients required for the clinical trial.

[0097] Furthermore, for example, by changing the conditions for each item included in the selection criteria based on priority and extracting the number of patients required for the clinical trial, it is possible to extract patients who are eligible for the clinical trial while optimizing the combination of items whose conditions are changed.

[0098] Furthermore, by predicting the number of patients who are likely to be suitable for the clinical trial at a given time based on the estimated results of the condition of patients extracted based on partially modified selection conditions, it is possible to select patients who are likely to be suitable for the clinical trial from a wider range. Furthermore, by extracting patients by partially modifying the selection conditions based on the patient's condition at a given time, it becomes easier to select the number of patients required for the clinical trial based on the estimated results of the patient's condition at the time of the clinical trial. Furthermore, by referring to the information on the suitability of patients extracted based on the modified conditions and the predicted results of the number of patients output by the selection support device 10, decision-making regarding the clinical trial can be easily made.

[0099] Each process in the selection support device 10 may be distributed and executed among multiple information processing devices connected via a network. For example, the process related to extracting patients who meet the selection conditions and the process related to calculating the degree of suitability for the selection conditions may be performed in different information processing devices. Furthermore, for example, the process related to extracting patients who meet the selection conditions, the process related to estimating the patient's condition at a predetermined time point, and the process related to predicting the number of patients who meet the selection conditions at a predetermined time point may be performed in different information processing devices. For example, the process in the extraction unit 12 and the processes in the estimation unit 13, the suitability calculation unit 14, and the prediction unit 15 may be performed in different information processing devices. It may be appropriately determined which information processing device performs each process in the selection support device 10.

[0100] Each process in the selection support device 10 can be realized by executing a computer program on a computer. Fig. 11 shows an example of the configuration of a computer 100 that executes a computer program that performs each process in the selection support device 10. The computer 100 includes a CPU (Central Processing Unit) 101, a memory 102, a storage device 103, an input / output I / F (Interface) 104, and a communication I / F 105.

[0101] The CPU 101 reads and executes computer programs for performing each process from the storage device 103. The CPU 101 may be configured as a combination of multiple CPUs. The CPU 101 may also be configured as a combination of a CPU and another type of processor. For example, the CPU 101 may be configured as a combination of a CPU and a graphics processing unit (GPU). The memory 102 is configured with a dynamic random access memory (DRAM) or the like, and temporarily stores the computer programs executed by the CPU 101 and data being processed. The storage device 103 stores the computer programs executed by the CPU 101. The storage device 103 is configured with, for example, a non-volatile semiconductor storage device. Other storage devices such as a hard disk drive may also be used for the storage device 103. The input / output I / F 104 is an interface that accepts input from a worker and outputs display data, etc. The communication I / F 105 is an interface that transmits and receives data between the terminal device 20, the client terminal device 30, the data management device 40, and other information processing devices. Furthermore, the terminal device 20 , the client terminal device 30 and the data management device 40 may also have the same configuration as the computer 100 .

[0102] The computer program used to execute each process can also be stored and distributed on a computer-readable recording medium that non-temporarily stores data. Examples of recording media that can be used include magnetic tapes for recording data and magnetic disks such as hard disks. Optical disks such as CD-ROMs (Compact Disc Read Only Memory) can also be used as recording media. Non-volatile semiconductor storage devices can also be used as recording media.

[0103] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0104] [Supplementary Note 1] A selection support device comprising: an acquisition means for acquiring selection conditions, which are conditions for selecting patients to be clinical trial subjects; an extraction means for extracting patients who meet the selection conditions based on data related to treatment; an estimation means for estimating the condition of each extracted patient at a predetermined time point based on the data related to treatment; a prediction means for predicting the number of patients who meet the selection conditions at the predetermined time point based on the estimated condition of each patient; and an output means for outputting the predicted number of patients.

[0105] [Supplementary Note 2] The selection support device described in Supplementary Note 1, wherein the prediction means predicts the number of patients who meet the selection conditions at the specified time point and a range of variation from the predicted value of the number of patients based on the estimated condition of each patient, and the output means outputs the predicted number of patients in association with the predicted range of variation.

[0106] [Supplementary Note 3] The selection support device according to Supplementary Note 1 or 2, wherein the predetermined time point is the start of a clinical trial or a time point during the clinical trial.

[0107] [Supplementary Note 4] The selection support device according to any one of Supplementary Notes 1 to 3, wherein the extraction means extracts patients who meet a condition obtained by changing a part of the selection condition based on the data related to the treatment.

[0108] [Supplementary Note 5] The selection support device according to any one of Supplementary Notes 1 to 3, wherein the extraction means extracts patients who meet conditions obtained by changing part of the selection conditions based on the results of estimating the condition of each extracted patient at the predetermined time point.

[0109] [Supplementary Note 6] The selection support device according to any one of Supplementary Notes 1 to 5, wherein the output means outputs statistical data relating to the condition of the patient who meets the selection condition at the predetermined time point.

[0110] [Supplementary Note 7] The selection support device according to any one of Supplementary Notes 1 to 6, wherein the extraction means extracts patients who meet the selection conditions based on data relating to treatment that has been pseudonymized.

[0111] [Supplementary Note 8] The selection support device according to any one of Supplementary Notes 1 to 5, wherein the output means outputs the state at the predetermined time point estimated by the estimation means for each patient who meets the selection conditions.

[0112] [Supplementary Note 9] The selection support device according to any one of Supplementary Notes 1 to 8, wherein the estimation means estimates the patient's condition at a predetermined time point from the patient attributes extracted by the extraction means and data related to the patient's treatment, using an estimation model generated by machine learning.

[0113] [Supplementary Note 10] The selection support device according to any one of Supplementary Notes 1 to 9, wherein the data relating to the treatment of the patient is time-series data relating to the treatment of the patient.

[0114] [Supplementary Note 11] A selection support method comprising: acquiring selection conditions that are conditions for selecting patients to be subject to a clinical trial; extracting patients who meet the selection conditions based on data related to treatment; estimating the condition of each extracted patient at a predetermined time point based on the data related to treatment; predicting the number of patients who meet the selection conditions at the predetermined time point based on the estimated condition of each patient; and outputting the predicted number of patients.

[0115] [Supplementary Note 12] The selection support method according to Supplementary Note 11, further comprising predicting the number of patients who meet the selection conditions at the predetermined time point and a range of variation from the predicted value of the number of patients based on the estimated condition of each patient, and outputting the predicted number of patients in association with the predicted range of variation.

[0116] [Supplementary Note 13] The selection support method according to Supplementary Note 11 or 12, wherein the predetermined time point is the start of a clinical trial or a time point during the clinical trial.

[0117] [Supplementary Note 14] The selection support method according to any one of Supplementary Notes 11 to 13, further comprising extracting patients who meet a condition obtained by changing a part of the selection condition based on the data related to the treatment.

[0118] [Supplementary Note 15] The selection support method according to any one of Supplementary Notes 11 to 13, wherein patients who meet conditions obtained by changing part of the selection conditions are extracted based on the results of estimating the condition of each extracted patient at the specified time point.

[0119] [Supplementary Note 16] The selection support method according to any one of Supplementary Notes 11 to 15, further comprising outputting statistical data relating to the conditions of a plurality of patients who meet the selection conditions at the predetermined time point.

[0120] [Supplementary Note 17] The selection support method according to any one of Supplementary Notes 11 to 16, wherein patients who meet the selection conditions are extracted based on pseudonymized data on treatment.

[0121] [Supplementary Note 18] The selection support method according to any one of Supplementary Notes 11 to 15, further comprising outputting an estimated state at the predetermined time point for each patient who meets the selection conditions.

[0122] [Supplementary Note 19] The selection support method according to any one of Supplementary Notes 11 to 18, wherein an estimation model generated by machine learning is used to estimate the patient's condition at a predetermined time point from the extracted patient attributes and data related to the patient's treatment.

[0123] [Supplementary Note 20] The selection support method according to any one of Supplementary Notes 11 to 19, wherein the data relating to the patient's treatment is time-series data relating to the patient's treatment.

[0124] [Supplementary Note 21] A recording medium that non-temporarily records a selection support program that causes a computer to execute the following processes: a process of acquiring selection conditions, which are conditions for selecting patients to be clinical trial subjects; a process of extracting patients who meet the selection conditions based on data related to treatment; a process of estimating the condition of each extracted patient at a predetermined time point based on the data related to treatment; a process of predicting the number of patients who meet the selection conditions at the predetermined time point based on the estimated condition of each patient; and a process of outputting the predicted number of patients.

[0125] [Appendix 22] A recording medium according to Appendix 21, on which a selection support program is non-temporarily recorded, which causes a computer to execute the following processes: a process of predicting the number of patients who meet the selection conditions at the predetermined time point and the range of fluctuation from the predicted value of the number of patients based on the estimated condition of each patient; and a process of outputting the predicted number of patients in association with the range of fluctuation of the prediction.

[0126] [Supplementary Note 23] The recording medium according to Supplementary Note 21 or 22, on which a selection support program is non-temporarily recorded, wherein the predetermined time point is the start of the clinical trial or a time point during the clinical trial.

[0127] [Appendix 24] A recording medium according to any one of Appendices 21 to 23, non-temporarily recording a selection support program that causes a computer to execute a process of extracting patients who meet conditions obtained by changing part of the selection conditions based on the data related to the treatment.

[0128] [Supplementary Note 25] A recording medium according to any one of Supplementary Notes 21 to 23, non-temporarily recording a selection support program that causes a computer to execute a process of extracting patients who meet conditions obtained by changing part of the selection conditions based on the results of estimating the condition of each extracted patient at the predetermined time point.

[0129] [Supplementary Note 26] A recording medium according to any one of Supplementary Notes 21 to 25, non-temporarily recording a selection support program that causes a computer to execute a process of outputting statistical data on the conditions of a plurality of patients who meet the selection conditions at the predetermined time point.

[0130] [Appendix 27] A recording medium according to any one of Appendices 21 to 26, non-temporarily recording a selection support program that causes a computer to execute a process of extracting patients who meet the selection conditions based on pseudonymized data on treatment.

[0131] [Appendix 28] A recording medium according to any one of Appendices 21 to 25, non-temporarily recording a selection support program that causes a computer to execute a process of outputting the estimated condition at the predetermined time point for each patient who meets the selection conditions.

[0132] [Appendix 29] A recording medium according to any one of Appendices 21 to 28, non-temporarily recording a selection support program that causes a computer to execute the following process: using an estimation model generated by machine learning to estimate a patient's condition at a predetermined time from extracted patient attributes and data related to the patient's treatment.

[0133] [Supplementary Note 30] The recording medium according to any one of Supplementary Notes 21 to 29, on which a selection support program is non-temporarily recorded, wherein the data relating to the patient's treatment is time-series data relating to the patient's treatment.

[0134] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0135] REFERENCE SIGNS LIST 10 Selection support device 11 Acquisition unit 12 Extraction unit 13 Estimation unit 14 Goodness of fit calculation unit 15 Prediction unit 16 Output unit 17 Storage unit 20 Terminal device 30 Client terminal device 40 Data management device 100 Computer 101 CPU 102 Memory 103 Storage device 104 Input / output I / F 105 Communication I / F

Claims

1. A means of obtaining selection criteria, which are the conditions for selecting patients to be included in a clinical trial, An extraction means for extracting patients who meet the selection criteria based on treatment data, Based on the data relating to the aforementioned treatment, an estimation means is provided to estimate the condition of each extracted patient at a predetermined point in time. A prediction means that predicts the number of patients who meet the selection criteria at a predetermined time based on the estimated condition of each patient, An output means for outputting the predicted number of patients and A selection support device equipped with the following features.

2. The prediction means predicts, based on the estimated condition of each patient, the number of patients who meet the selection criteria at the predetermined time point, and the range of variation from the predicted value of the number of patients. The output means outputs the predicted number of patients in relation to the range of variation in the prediction. The selection support device according to claim 1.

3. The aforementioned predetermined time is the start of the clinical trial or a time during the clinical trial. The selection support device according to claim 1 or 2.

4. The extraction means extracts patients who meet conditions that have been modified from the selection conditions based on the data relating to the treatment. The selection support device according to claim 1 or 2.

5. The extraction means extracts patients who meet conditions that have been modified from the selection conditions, based on the estimated state of each extracted patient at the predetermined time point. The selection support device according to claim 1 or 2.

6. The output means outputs statistical data relating to the status of multiple patients that meet the selection criteria at a predetermined time point. The selection support device according to claim 1 or 2.

7. The extraction means extracts patients who meet the selection criteria based on the treatment data that has been anonymized. The selection support device according to claim 1 or 2.

8. The output means outputs the state at the predetermined time point estimated by the estimation means for each patient that meets the selection criteria. The selection support device according to claim 1 or 2.

9. We obtained the selection criteria, which are the conditions for selecting patients to be included in the clinical trial. Based on treatment data, patients who meet the above selection criteria are extracted. Based on the data related to the aforementioned treatment, the condition of each extracted patient at a predetermined point in time is estimated. Based on the estimated condition of each patient, the number of patients who meet the selection criteria at the predetermined time is predicted. Output the predicted number of patients. Selection support method.

10. The process of obtaining selection criteria, which are the conditions for selecting patients to be included in the clinical trial, Based on treatment data, a process is performed to extract patients who meet the aforementioned selection criteria, Based on the data related to the aforementioned treatment, a process is performed to estimate the condition of each extracted patient at a predetermined point in time, A process to predict the number of patients who meet the selection criteria at a predetermined time, based on the estimated condition of each patient, Process to output the predicted number of patients and A selection support program that has a computer perform the necessary actions.