Clinical trial support apparatus, clinical trial support method, and clinical trial support program

Abstract

To provide a clinical trial support apparatus capable of efficiently selecting patients being clinical trial targets.SOLUTION: A clinical trial support apparatus includes an acquisition unit 11, a complementing unit 12, a selection unit 13, and an output unit 15. The acquisition unit 11 acquires data related to treatment of patients. The complementing unit 12 complements, among data used for selecting patients being clinical trial targets, data that is missing in the data related to the treatment. The selection unit 13 selects patients being clinical trial targets on the basis of the complemented data. The output unit 15 outputs information related to the selected patients being clinical trial targets. With such a configuration, the clinical trial support apparatus 10 can support decision-making regarding selection of patients being clinical trial targets.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a clinical trial support device and the like. [Background technology]

[0002] In clinical trials of pharmaceuticals, the person in charge of selecting candidate patients for the clinical trial extracts patients who meet the selection criteria for clinical trial patients, for example, by checking the descriptions in medical records. For example, the person in charge determines whether the selection criteria match the contents of the medical records. Then, the person in charge extracts patients whose selection criteria match the contents of their medical records as candidate patients for the clinical trial. In addition, the person in charge must check the medical records of many patients to find patients who meet the selection criteria. For this reason, a system that supports the selection of candidate patients for clinical trial is sometimes used to select candidate patients for clinical trial.

[0003] The clinical trial candidate extraction device of Patent Document 1 stores information on the patient's illness and information on the investigational drug administered to the patient, and extracts clinical trial candidates based on the number of times the administered investigational drug has been changed. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-194595 Summary of the Invention [Problem to be solved by the invention]

[0005] The technology described in Patent Document 1 may require a lot of work to select patients for clinical trials.

[0006] In order to solve the above-mentioned problems, the present disclosure aims to provide a clinical trial support device etc. that can efficiently select patients to be included in clinical trials. [Means for solving the problem]

[0007] In order to solve the above problems, the clinical trial support device disclosed herein comprises an acquisition means for acquiring data related to patient treatment, a completion means for completing missing data related to treatment among data used to select patients for the clinical trial, a selection means for selecting patients for the clinical trial based on the completed data, and an output means for outputting information related to the selected patients for the clinical trial.

[0008] The clinical trial support method disclosed herein acquires data related to patient treatment, complements missing data related to treatment from among the data used to select patients for the clinical trial, selects patients for the clinical trial based on the complemented data, and outputs information related to the selected patients for the clinical trial.

[0009] The clinical trial support program disclosed herein causes a computer to perform the following processes: acquiring data related to patient treatment; complementing missing data related to treatment from among data used to select patients for the clinical trial; selecting patients for the clinical trial based on the complemented data; and outputting information related to the selected patients for the clinical trial. [Effects of the Invention]

[0010] According to the present disclosure, patients for clinical trials can be efficiently selected. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a clinical trial support system according to the present disclosure. [Figure 2] FIG. 1 is a diagram illustrating an example of the configuration of a clinical trial support device according to the present disclosure. [Figure 3] FIG. 1 illustrates an example of patient inclusion criteria for a clinical trial in the present disclosure. [Figure 4] FIG. 1 illustrates an example of patient exclusion criteria from a clinical trial in the present disclosure. [Figure 5]FIG. 10 illustrates an example of a graph showing patient relationships in accordance with the present disclosure. [Figure 6] FIG. 10 is a diagram showing an example of the extraction results of patients who are subject to clinical trials in the present disclosure. [Figure 7] FIG. 10 is a diagram showing an example of the extraction results of patients who are subject to clinical trials in the present disclosure. [Figure 8] FIG. 10 is a diagram illustrating an example of an operation flow of a clinical trial support device according to the present disclosure. [Figure 9] FIG. 1 is a diagram illustrating an example of the hardware configuration of a clinical trial support device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

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

[0013] A clinical trial support system is, for example, a system that supports the selection of patients to be subjects of clinical trials. Patients to be subjects of clinical trials are, for example, patients who are the subjects of clinical trials for new pharmaceuticals. The clinical trial support system outputs information about patients to be subjects of clinical trials, selected from patients receiving treatment at medical institutions, for example. Patients receiving treatment at medical institutions may include patients who were receiving treatment.

[0014] Patients selected as clinical trial subjects are, for example, patients with a condition suitable for verifying the effectiveness of a pharmaceutical product. For example, patients selected as clinical trial subjects are patients whose treatment-related data meet the selection criteria for clinical trial patients. The selection criteria are conditions for selecting a patient as a clinical trial patient. The selection criteria include, for example, inclusion criteria and exclusion criteria. The inclusion criteria indicate, for example, the criteria for including a patient in a clinical trial. Furthermore, the exclusion criteria indicate, for example, the criteria for excluding a patient from a clinical trial. In other words, the exclusion criteria are, for example, the criteria for not selecting a patient as a clinical trial patient.

[0015] When selecting patients for clinical trials, even patients with similar symptoms may not be included as candidates because some of the data necessary for selecting patients for clinical trials is missing. Furthermore, for example, in clinical trials of pharmaceuticals for diseases and injuries with few cases, it may be difficult to select patients suitable for the clinical trial because the number of patients who can be candidates for selection is small. Clinical trial support systems improve the efficiency of selecting patients for clinical trials by, for example, complementing missing data among the treatment-related data necessary for selecting patients for clinical trials, thereby expanding the range of patients eligible for selection. Furthermore, for example, for diseases and injuries with few cases, complementing missing data to select patients for clinical trials may improve the likelihood of selecting the required number of patients for the clinical trial.

[0016] Here, an example of the configuration of the clinical trial support device 10 will be described. Fig. 2 shows an example of the configuration of the clinical trial support device 10. The clinical trial support device 10 basically comprises an acquisition unit 11, a complementation unit 12, a selection unit 13, and an output unit 15. The clinical trial support device 10 may further comprise, for example, a prediction unit 14 and a storage unit 16.

[0017] The acquisition unit 11 acquires data related to patient treatment. For example, the acquisition unit 11 acquires data related to patient treatment for each patient who may be a subject of a clinical trial. The data related to patient treatment includes, for example, patient attributes and patient treatment data.

[0018] Patient attributes are, for example, information indicating the characteristics of a patient that relates to the patient's condition and that does not change with treatment. Patient attributes are, for example, information on one or more of the patient's gender, age, family structure, family medical history, nationality, and race. Patient attributes are not limited to the above.

[0019] Patient treatment data is, for example, a record of medical procedures performed on a patient. The patient treatment data is, for example, a record of one or more items of diagnosis, examination, medication, surgery, follow-up, and patient condition. The patient condition includes, for example, one or more items of disease information, complication information, biomarkers, disease status, guideline score, treatment effect, and test results. The guideline score is, for example, an index indicating the risk of each injury or illness. For example, the risk of each injury or illness is an index indicating the likelihood that a patient will contract or be affected by the injury or illness. For example, the guideline score is calculated as an index indicating the extent to which data regarding the patient's treatment matches the criteria established for each injury or illness. Treatment effect is, for example, the effect of treatment, drug administration, and surgery. The effect is not limited to the above. Test results are, for example, the results of biopsies, diagnostic imaging tests, and genomic tests. The test results are not limited to the above. The patient treatment data may also include information about the person who performed the medical procedure. Information about the person who performed the medical procedure is, for example, information indicating the medical professional who performed the medical procedure on the patient. The information indicating the medical professionals who performed medical procedures on the patient is, for example, the names or identifiers of doctors, nurses, pharmacists, and physical therapists.

[0020] Patient treatment data is recorded as data for each patient in, for example, an electronic medical record. It may be patient treatment data or test data. Patient treatment data may also be data written on medical receipts. Patient treatment data is not limited to the above. Treatment-related data is not limited to the above.

[0021] The acquisition unit 11 acquires data related to the treatment of a patient, for example, as structured data. Structured data is data from which data can be extracted according to rules, for example. For example, in structured data, data items are associated with the data in each item. In structured data, for example, by specifying a data item, it is possible to extract data associated with that item. A data item is information indicating what type of data each item contains. For example, if the data is the name of an illness or injury suffered by the patient, the item is the name of the illness or injury.

[0022] The acquiring unit 11 may acquire data related to the treatment of a patient as unstructured data. The unstructured data is, for example, data in an unstructured state. The unstructured data is, for example, data in which a medical professional describes the patient's condition in writing. The medical professional is, for example, a doctor. The medical professional may be, for example, a nurse, a pharmacist, a psychotherapist, or a physical therapist. The medical professional is not limited to the above. The unstructured data may also be image data. The image data is, for example, imaging data from an X-ray examination, an endoscopic examination, a CT (Computed Tomography) examination, or an MRI (Magnetic Resonance Imaging) examination. The image data is not limited to the above. The acquiring unit 11 acquires data related to the treatment of a patient from, for example, the data management device 30.

[0023] The acquisition unit 11 further acquires, for example, selection criteria for patients to be subject to the clinical trial. For example, the acquisition unit 11 acquires at least one of criteria for inclusion of patients in the clinical trial and criteria for exclusion from the clinical trial as the selection criteria. The selection criteria include, for example, multiple criteria used to select patients to be subject to the clinical trial. The acquisition unit 11 acquires the selection criteria for patients to be subject to the clinical trial from, for example, the terminal device 20.

[0024] The inclusion criteria is, for example, information indicating the conditions for patients to be eligible for the clinical trial. The conditions for patients to be eligible for the clinical trial are indicated using the attributes of patients suitable as clinical trial patients and medical records of their treatment. The exclusion criteria is information indicating the conditions for excluding patients from the patients to be eligible for the clinical trial. That is, the exclusion criteria is, for example, information indicating the conditions for patients not to be selected as clinical trial patients. The exclusion criteria is indicated using the attributes of patients not suitable as clinical trial patients and medical records of their treatment.

[0025] Figures 3 and 4 are examples of selection criteria for conducting clinical trials of lung cancer therapeutic drugs. Figure 3 is an example of a sentence indicating the criteria for patient inclusion in the clinical trial, among the selection criteria. In the example of the inclusion criteria for the clinical trial in Figure 3, the inclusion criteria include multiple criteria. Also, Figure 4 is an example of a sentence indicating the criteria for patient exclusion from the clinical trial, among the selection criteria. In the example of the exclusion criteria for the clinical trial in Figure 4, the exclusion criteria include multiple criteria.

[0026] The acquiring unit 11 may also acquire conditions to be prioritized in selecting patients for clinical trial subjects. The prioritized conditions may be conditions indicating criteria included in the selection criteria. For example, when selecting patients by prioritizing conditions related to age among the criteria included in the selection criteria, the acquiring unit 11 acquires information indicating that age is to be prioritized in selection. When selecting patients by prioritizing age, the acquiring unit 11 may also acquire information indicating, for example, an age to be prioritized among the ages indicated as criteria in the selection criteria. The acquiring unit 11 may also acquire the priority of each criterion included in the selection criteria as a condition to be prioritized in selecting patients for clinical trial subjects. The priority is, for example, an index indicating the degree to which each criterion included in the selection criteria is prioritized in selecting patients. The acquiring unit 11 acquires the conditions to be prioritized in selecting patients for clinical trial subjects from, for example, the terminal device 20.

[0027] The complementing unit 12 complements missing data related to treatment among the data used to select patients for clinical trial. For example, the complementing unit 12 complements missing data related to treatment among the data used to calculate the degree of conformance of the selection criteria for patients for clinical trial. For example, the complementing unit 12 compares the selection criteria with the data related to the treatment of each patient and extracts data items that need to be complemented. Then, the complementing unit 12 complements, for example, the missing data for the extracted patients that need to be complemented.

[0028] The completion unit 12 completes missing data in the treatment-related data, for example, by using a graph showing the relationships between patients. The graph showing the relationships between patients includes, for example, nodes corresponding to the treatment-related data of each patient and edges corresponding to lines connecting patients whose treatment-related data are similar. The graph showing the relationships between patients is generated, for example, based on the patient treatment-related data. The process of generating the graph showing the relationships between patients will be described later.

[0029] The completion unit 12 completes missing data related to treatment by using data of patients whose data related to treatment is similar to that of a patient whose data related to treatment needs to be completed, based on, for example, a graph showing the relationships between patients. The completion unit 12 completes data related to a patient whose data needs to be completed by using data related to treatment of a patient connected by an edge to the patient whose data needs to be completed in the graph showing the relationships between patients.

[0030] For example, the complementing unit 12 complements data that needs to be complemented using data of the same item among data related to the treatment of a patient connected by an edge. For example, if liver function data is missing from blood test data among the data related to the treatment, the complementing unit 12 complements the missing data using data on the liver function test results of a patient connected by an edge to the patient with missing data.

[0031] When a patient to be complemented is connected to multiple patients via edges, the complementing unit 12 may complement missing data using data from the multiple patients connected via edges. For example, when a patient to be complemented is connected to multiple patients via edges, the complementing unit 12 complements missing data so that the data is the average of the multiple patients connected via edges. When data is missing from a patient connected via edges, the complementing unit 12 may complement the missing data using data from a patient further connected via edges to the patient connected via edges. The complementing unit 12 may also complement the missing data by weighting based on the hierarchy of the edges. The hierarchy of the edges is, for example, the number of edges between patients. In this case, the complementing unit 12 weights the data of patients in a hierarchy closer to the patient to be complemented so that it is weighted more heavily. When a patient to be complemented is connected to multiple patients via edges, the complementing unit 12 may complement the missing data using data from a patient who has the highest similarity in treatment data to the patient to be complemented, among the multiple patients connected via edges. The complementing unit 12, for example, generates a graph showing the relationships between patients based on the treatment data of each patient. Of the treatment-related data, the data used to generate the graph is set, for example, according to selection criteria. The completion unit 12 generates a graph showing the relationships between patients, for example, by the following process. For example, the completion unit 12 converts the treatment-related data for each patient into an embedding vector using a language model. The language model converts the treatment-related data into an embedding vector using, for example, a dictionary. Then, the completion unit 12 converts the treatment-related data into a feature vector using, for example, the converted embedding vector. In this case, the feature vector is a multidimensional vector that reflects the treatment-related data for each patient. In other words, the feature vector in this case is a multidimensional vector that represents the features of each patient.

[0032] The language model converts the data related to each patient's treatment into an embedding vector using, for example, a dictionary that covers the medical field in general. The language model may convert the data related to each patient's treatment into an embedding vector using multiple dictionaries. For example, the language model converts the data related to each patient's treatment into an embedding vector using a dictionary that covers the medical field in general and a dictionary that covers the medical department in which the drug under clinical trial is used.

[0033] For example, Word2Vec can be used as the language model. For example, GPT-2 (Generative Pre-trained Transformer-2), GPT-3, GPT-3.5, or GPT-4 can be used as the language model. In addition, T5 (Text-to-Text Transfer Transformer), BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (Robustly optimized BERT approach), or ELECTRA (Efficiently Learning an Encoder that Classifies Token Replacements Accurately) can also be used as the language model. The language model used for conversion to an embedding vector is not limited to the above.

[0034] The completion unit 12 calculates the similarity between patients, for example, using the distance between the feature vectors of each patient. The completion unit 12 calculates the similarity between patients, for example, by calculating the distance between the feature vectors. The distance between the vectors is, for example, Euclidean distance. The distance between the feature vectors is not limited to Euclidean distance. Alternatively, the completion unit 12 may calculate the similarity between patients by calculating the cosine similarity between the feature vectors. The completion unit 12 generates a graph showing the relationship between patients by connecting nodes with edges based on the similarity between patients. For example, the completion unit 12 generates a graph by connecting nodes of patients with edges between patients whose similarity is equal to or greater than a predetermined criterion. The predetermined criterion is set, for example, so that patients are considered to be similar in terms of conducting a clinical trial when the similarity exceeds the predetermined criterion. For example, patients being similar in terms of conducting a clinical trial means that the effects of a drug being tested are expected to be similar between patients. Here, the nodes represent, for example, each patient. The edges represent, for example, the relationship between patients. That is, a graph showing the relationships between patients is, for example, a graph in which nodes representing each patient are connected by edges connecting similar patients. Similarity means, for example, that data related to treatments have a relationship. Relational data related to treatments means, for example, that if one meets the selection criteria for a clinical trial, there is a high probability that the other will also meet the selection criteria for the clinical trial. In other words, similarity means, for example, that if one patient meets certain selection criteria, there is a high probability that the other patient will also meet the criteria. Similarity can also include identity.

[0035] FIG. 5 is a diagram schematically illustrating an example of a graph showing relationships between patients generated by the complementing unit 12. In the example of FIG. 5, the graph showing relationships between patients includes "Patient A," "Patient B," "Patient C," and "Patient D" as nodes. In the example of FIG. 5, "Patient A" is connected to "Patient B" and "Patient C" by edges. In the example of FIG. 5, "Patient D" is connected to "Patient B" and "Patient C" by edges. In the example of FIG. 5, data corresponding to selection criteria among the treatment-related data for each patient corresponding to a node is shown. In the example of FIG. 5, selection criterion "1" is a criterion related to the name of the illness or injury. In the example of FIG. 5, selection criterion "2" is a criterion related to the patient's age. In the example of FIG. 5, selection criterion "3" is test data for the white blood cell count in the blood. In the example of FIG. 5, for example, test data for the white blood cell count for "Patient A" is missing. When data is missing as shown in the example of FIG. 5, the complementing unit 12 complements the white blood cell count data of "Patient A" using the test data of the white blood cell counts of "Patient B" and "Patient C" that are connected to "Patient A" by edges. For example, the complementing unit 12 complements the white blood cell count data of "Patient A" by using the average value of the white blood cell counts of "Patient B" and "Patient C" that are connected to "Patient A" by edges. The complementing unit 12 may also complement the white blood cell count data of "Patient A" by using the white blood cell count value of a patient that is most similar to "Patient A" among "Patient B" and "Patient C."

[0036] The completion unit 12 may generate a graph showing the relationships between patients using a graph generation model. The graph generation model is, for example, a machine learning model that generates a graph showing the relationships between patients from data related to the treatment of each patient. For example, the completion unit 12 converts each piece of treatment-related data for each patient into an embedding vector using a language model. Then, the completion unit 12 generates a graph showing the relationships between patients, for example, using the embedding vector and the graph generation model as input. The graph generation model is generated, for example, by learning the relationship between the data related to the treatment of each patient and the graph showing the relationships between patients. The graph generation model is generated, for example, by deep learning using a neural network. The learning algorithm used to generate the graph generation model is not limited to the above. The graph generation model may also be generated, for example, in a device external to the clinical trial support device 10.

[0037] The completion unit 12 may complete missing data using unstructured data. When the unstructured data is text recorded in an electronic medical record, for example, the completion unit 12 completes missing data from the text recorded in the electronic medical record. When the unstructured data is text recorded in the electronic medical record, the process of completing missing data using unstructured data is performed, for example, as follows. For example, the completion unit 12 identifies data missing in the selection of patients to be included in the clinical trial. Then, for example, the completion unit 12 extracts a description corresponding to the missing data from the electronic medical record using a language model based on information indicating the missing data. For example, the language model can be GPT-2 (Generative Pre-trained Transformer-2), GPT-3, GPT-3.5, or GPT-4. Furthermore, the language model may be T5 (Text-to-Text Transfer Transformer), BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (Robustly optimized BERT approach), or ELECTRA (Efficiently Learning an Encoder that Classifies Token Replacements Accurately). The language model used to extract missing data is not limited to the above. The language model used to extract missing data may be the same as the language model that converts treatment-related data into embedding vectors. The language model used to extract missing data may be a different language model from the language model that converts treatment-related data into embedding vectors.

[0038] Furthermore, when the unstructured data is image data, the completion unit 12 completes missing data, for example, using a diagnosis result from an image diagnostic model. The image diagnostic model is, for example, a machine learning model that estimates the name of an injury or disease using image data captured during an examination as input. The image diagnostic model may be a machine learning model that estimates the size of a lesion using image data captured during an examination as input. The image diagnostic model may be a machine learning model that estimates the name of an injury or disease and the size of a lesion using image data captured during an examination as input. For example, when data on the size of a lesion in a stomach cancer patient is missing, the completion unit 12 estimates the size of the lesion in the stomach using the image diagnostic model based on image data from an endoscopic examination of the stomach.

[0039] The image diagnostic model is generated, for example, by learning the relationship between image data captured during an examination and the name of an injury or disease. The image diagnostic model may also be generated by learning the relationship between image data captured during an examination and the extent of a lesion on the image data. The image diagnostic model may also be generated by learning the relationship between image data captured during an examination and the name of an injury or disease and the extent of a lesion on the image data. The image diagnostic model is generated, for example, by deep learning using a neural network. The image diagnostic model is generated, for example, in a system external to the clinical trial support device 10.

[0040] When the selection unit 13 selects patients to be included in the clinical trial based on data on past treatments, the completion unit 12 may complete missing data from the data on past treatments. For example, the completion unit 12 completes missing data for patients for whom data on past treatments covering the period during which the clinical trial is conducted is recorded. When the prediction unit 14 predicts data on treatments, the completion unit 12 may complete missing data in the data on treatments predicted by the prediction unit 14. For example, the prediction unit 14 completes missing data from predicted values ​​of data on the clinical trial.

[0041] The completion unit 12 may generate data for each patient that can be considered to be in a state in which missing data has been completed, as data in which missing data has been completed. For example, the completion unit 12 generates feature vectors for each patient as data in which missing data has been completed, based on a graph showing the relationships between patients generated based on data on the patients' treatments and the selection criteria for patients to be included in the clinical trial. For example, the completion unit 12 generates feature vectors for each patient based on the graph showing the relationships between patients and the conformance of the treatment-related data to the selection criteria. In this case, the completion unit 12 calculates, for example, the conformance of the treatment-related data to the selection criteria. For example, the completion unit 12 calculates the ratio of the number of criteria that the treatment-related data satisfies to the number of criteria included in the selection criteria as the conformance of the treatment-related data to the selection criteria.

[0042] The completion unit 12 converts, for example, a graph showing the fitness for each criterion included in the selection criteria and the relationships between patients into a feature vector for each patient. The completion unit 12 converts, for example, a graph showing the fitness for the selection criteria and the relationships between patients into a feature vector for each patient using a conversion model that converts the graph showing the fitness for each criterion included in the selection criteria and the relationships between patients into a single feature vector. The conversion model is, for example, a machine learning model that converts the graph showing the fitness for each criterion included in the selection criteria and the relationships between patients into a feature vector for each patient.

[0043] The conversion model is generated, for example, as follows. In the first stage of generating the conversion model, the learning device that generates the conversion model learns a vector representation for each piece of information related to attributes and treatments, for example, using a message passing technique in which messages are exchanged via edges in a graph showing the relationships between patients. In the second stage, the learning device learns a vector representation for each node that combines vectors for each piece of information related to fitness, attributes, and treatments. The nodes correspond, for example, to each patient. The conversion model generated in this manner can convert the graph showing the fitness and the relationships between patients into a feature vector. This learning technique is also called embedding propagation. The conversion model is generated, for example, in a system external to the clinical trial support device 10. The conversion model may also be generated by a learning means (not shown) included in the clinical trial support device 10.

[0044] The selection unit 13 selects patients to be included in the clinical trial based on the supplemented data. The selection unit 13 selects patients to be included in the clinical trial based on, for example, the suitability of the treatment data for each patient with respect to the selection criteria. Furthermore, for patients with no missing data, the selection unit 13 performs a process of selecting patients to be included in the clinical trial using, for example, treatment data that has not been subjected to supplementation processing. In other words, the selection unit 13 selects patients to be included in the clinical trial based on treatment data for patients whose data has been supplemented and treatment data for patients whose data does not require supplementation.

[0045] As a process for selecting patients to be included in the clinical trial, the selection unit 13, for example, calculates the degree of suitability of the treatment-related data of each patient with respect to the selection criteria for patients to be included in the clinical trial. In this case, for example, for patients whose data has been supplemented, the selection unit 13 calculates the degree of suitability with respect to the selection criteria using the supplemented treatment-related data. The degree of suitability is, for example, the ratio of the number of criteria included in the selection criteria to the number of criteria for which the treatment-related data satisfies the conditions. The selection unit 13 may calculate the degree of suitability by weighting each of the criteria included in the selection criteria. Then, for example, the selection unit 13 selects patients whose degree of suitability meets or exceeds the standard as patients to be included in the clinical trial. The suitability criteria for selecting patients to be included in the clinical trial are set so that patients whose degree of suitability exceeds the standard are suitable patients to be included in the clinical trial.

[0046] The selection unit 13 may extract multiple levels of suitability of patients for a clinical trial. By setting multiple levels of suitability, it becomes possible to extract patients who are candidates for clinical trial subjects, for example, when some of the criteria are relaxed. The selection unit 13, for example, extracts information indicating which of multiple levels of suitability for the clinical trial estimated by the extraction model for each patient. The selection unit 13 may also extract the number of patients who are suitable for the clinical trial. Furthermore, when multiple levels according to suitability for the clinical trial are set, the selection unit 13 may extract the number of patients for each level.

[0047] The selection unit 13 may select patients in the trial group and patients in the control group as patients to be included in the clinical trial. Patients in the trial group are, for example, patients who are administered the investigational drug. Patients in the control group are, for example, patients who are not administered the investigational drug for comparison with patients in the trial group. Patients in the control group are, for example, administered a substance called a placebo. A placebo is, for example, a substance that does not contain an active ingredient and is indistinguishable from the investigational drug.

[0048] The selection unit 13 may select clinical trial patients further based on data on the treatment of patients already selected as clinical trial patients. The selection unit 13, for example, selects clinical trial patients based on data on the treatment of the selected clinical trial patients so as to avoid bias in the selected clinical trial patients. "No bias" means, for example, that there is no bias in the treatment data between patients in the control group and patients in the clinical trial group. For example, the selection unit 13 randomly selects patients in the clinical trial group and patients in the control group from patients whose treatment data meet the selection criteria.

[0049] The selection unit 13 may select patients from either the control group or the clinical trial group as patients to be treated in the clinical trial. For example, when patients from the clinical trial group have already been selected, the selection unit 13 selects patients from the control group. In this case, the selection unit 13 selects patients from the control group so as to avoid bias between patients from the clinical trial group and patients from the control group. For example, the selection unit 13 selects patients from the control group based on data on patients from the clinical trial group that have already been selected as patients to be treated in the clinical trial. For example, when patients from the selected clinical trial group are in their 20s and 30s, the selection unit 13 preferentially selects patients in their 20s and 30s as patients to be treated in the control group. The selection unit 13 may select some patients from the control group based on data on patients from the clinical trial group that have already been selected as patients to be treated in the clinical trial. The selection unit 13 may select patients from the control group based on data on patients from the control group that have already been selected as patients to be treated in the clinical trial.

[0050] The selection unit 13 may further select patients for the clinical trial group based on data of patients in the clinical trial group who have already been selected as patients to be treated in the clinical trial. For example, the selection unit 13 further selects patients for the clinical trial group so as to avoid bias in data on treatment for patients within the criteria included in the selection criteria. For example, if the selection criteria include a condition that patients must be between 20 and 50 years old, and the patients in the selected clinical trial group are in their 20s and 30s, the selection unit 13 may preferentially select patients in their 40s as patients for the clinical trial group. The selection unit 13 may also further select patients for the control group based on data of patients in the control group who have already been selected as patients to be treated in the clinical trial.

[0051] When the completion unit 12 generates a feature vector for each patient, the selection unit 13 extracts, for example, patients suitable for the clinical trial as candidate patients for the clinical trial based on the feature vector for each patient. The feature vector for each patient is a feature vector converted by the completion unit 12 from a graph showing each patient's suitability for the selection criteria and their relationships. The selection unit 13, for example, extracts patients whose suitability for the clinical trial meets or exceeds a standard as patients suitable for the clinical trial. The standard for suitability for the clinical trial is set, for example, so that patients whose suitability exceeds the standard are suitable as clinical trial patients. If there is a condition for patients to be selected with priority, the selection unit 13 may select patients for the clinical trial by changing the corresponding criterion among the selection criteria so that patients are selected with priority. For example, if the selection criteria include a condition that patients must be between 20 and 50 years old, and the previously selected patients are in their 20s and 30s, the selection unit 13 may change the age criterion among the selection criteria to patients in their 40s to select patients for the clinical trial.

[0052] The selection unit 13 extracts candidate patients for clinical trial subjects using, for example, an extraction model. The extraction model is used, for example, to extract patients suitable for the clinical trial based on the feature vectors of each patient. The extraction model is, for example, a machine learning model that estimates the suitability for the clinical trial using the feature vectors of each patient as input. The suitability for the clinical trial is, for example, the probability that the patient will be suitable for the clinical trial. The selection unit 13 extracts patients suitable for the clinical trial based on, for example, the suitability for the clinical trial of each patient estimated by the extraction model. In other words, the selection unit 13 extracts, for example, patients whose suitability for the clinical trial of each patient estimated by the extraction model is equal to or exceeds a standard as patients suitable for the clinical trial.

[0053] The extraction model is generated, for example, by learning the relationship between each patient's feature vector and whether or not the patient is suitable for the clinical trial. The feature vector for each patient is a feature vector converted by the complementing unit 12 using a conversion model from a graph showing the suitability and the relationship between patients. The suitability for the clinical trial is actual data on whether or not each patient is suitable for the clinical trial criteria. The extraction model is generated, for example, by deep learning using a neural network. The extraction model is generated in a system of a device external to the clinical trial support device 10.

[0054] The selection unit 13 may select patients to be included in the clinical trial based on data related to past treatments. Alternatively, the selection unit 13 may select patients to be included in the clinical trial based on data obtained by supplementing missing data from the data related to past treatments. The selection unit 13 selects patients to be included in the clinical trial from patients who met the selection criteria at a past point in time and for whom data related to treatments covering the period during which the clinical trial was conducted has been recorded since the time the selection criteria were met. For example, the selection unit 13 selects patients to be included in the control group from among patients to be included in the clinical trial based on data related to past treatments. Meeting the selection criteria may mean, for example, that the degree of fit of the data related to the patient's treatment with the selection criteria for selecting patients to be included in the clinical trial satisfies the criteria for selection as a patient to be included in the clinical trial.

[0055] The prediction unit 14 predicts data related to a patient's treatment at a predetermined time point, for example, based on data related to the treatment. For example, the prediction unit 14 predicts the patient's future condition. The predetermined time point is, for example, the time point when a clinical trial begins. The predetermined time point may be the time point when patients for the clinical trial are selected. The predetermined time point may be a time point during the implementation of the clinical trial. The predetermined time point is not limited to the above. The prediction unit 14 predicts data related to treatment at a predetermined time point for a patient selected by the selection unit 13. Furthermore, the prediction result of the data related to treatment at a predetermined time point may be used as data for which the complementation unit 12 complements data. Furthermore, the prediction result of the data related to treatment at a predetermined time point may be used as data related to treatment for the selection unit 13 to extract patients for the clinical trial.

[0056] The prediction unit 14 predicts the treatment-related data at a predetermined time point using, for example, a prediction formula for each item of the treatment-related data. The prediction formula is, for example, a function with the predicted value of the treatment-related data as the objective variable, and the treatment-related data at a past time point or the present time and the elapsed time from the past time point or the present time as the explanatory variables.

[0057] The prediction unit 14 may use a prediction model to predict data related to a patient's treatment at a predetermined time point. The prediction model is, for example, a machine learning model that inputs data related to treatment at a past time point or the present time and the elapsed time from the past time point or the present time, and predicts data related to treatment at a predetermined time point. The prediction model is generated, for example, by learning the relationship between data related to treatment at a first time point, the time from the first time point to a second time point, and the data related to treatment at the second time point. The prediction model is generated, for example, by deep learning using a neural network. The machine learning algorithm used to generate the prediction model is not limited to the above. The prediction model may be generated, for example, in a device external to the clinical trial support device 10.

[0058] The output unit 15 outputs information about the selected patients who are clinical trial subjects. The output unit 15 outputs, for example, identification information about the patients who are clinical trial subjects selected by the selection unit 13 and data about the treatment of each patient. For example, the output unit 15 outputs information about the selected patients who are clinical trial subjects. For example, the output unit 15 outputs data about the treatment included in the selection criteria as data about the treatment of each patient.

[0059] The output unit 15 may output, as information about the selected patients who are the subjects of the clinical trial, the suitability for the clinical trial for each of the patients who are the subjects of the clinical trial selected by the selection unit 13. For example, the output unit 15 may output, as the suitability for the clinical trial, the suitability of the data related to the treatment of each patient with respect to the selection criteria.

[0060] The output unit 15 may output information about the selected clinical trial patients for each patient attribute. For example, the output unit 15 outputs information about the clinical trial patients for each patient age group. For example, if the selection criteria specify that the clinical trial patients must be aged 20 to 50, the output unit 15 outputs the number of clinical trial patients in their 20s, 30s, and 40s.

[0061] FIG. 6 is an example of a display screen displaying the results of extracting patients for clinical trials. On the display screen of FIG. 6, "patient names" indicating patients extracted as patients for clinical trials are associated with data on the treatment of each extracted patient. The "patient name" is, for example, a patient's identifier. The "patient name" may be, for example, the patient's name, identification number, or identification symbol. On the display screen of FIG. 6, the "patient name" is associated with the "name of illness," "age," and "gender" from among the data on the treatment of each patient. The "name of illness" is, for example, the name of the illness or injury being treated. The data on the treatment associated with the "patient name" is not limited to the above. By outputting the results of extracting patients for clinical trials as shown in the example display screen of FIG. 6 from the output unit 15, a person in charge of selecting patients for clinical trials can determine which patients to include in the clinical trial by referring to the condition of each patient.

[0062] FIG. 7 is an example of a display screen that displays the degree of suitability of each extracted patient for the selection criteria as a result of extracting patients for clinical trials. The display screen of FIG. 7 associates the "patient name" indicating the patient extracted as a patient for clinical trials with the degree of suitability of each extracted patient for the clinical trial. The "patient name" is, for example, a patient identifier. The "patient name" may be, for example, the patient's name, identification number, or identification symbol. By outputting the extraction result of patients for clinical trials as shown in the example display screen of FIG. 7 from the output unit 15, a person in charge of selecting patients for clinical trials can determine which patients to select by referring to the degree of suitability for the selection criteria.

[0063] When the prediction unit 14 predicts data related to treatment at a predetermined time point, the output unit 15 may output a predicted value of data related to treatment for the patient who is a subject of the clinical trial selected by the selection unit 13. For example, the output unit 15 outputs a predicted value of data related to treatment at the start of the clinical trial for the patient who is a subject of the clinical trial selected by the selection unit 13, based on the prediction result of the prediction unit 14.

[0064] The memory unit 16 stores, for example, data related to the process of selecting patients for clinical trial subjects. The memory unit 16 stores, for example, data related to the treatment of each patient acquired by the acquisition unit 11. The memory unit 16 stores, for example, information indicating patients for clinical trial subjects selected by the selection unit 13. The memory unit 16 stores, for example, selection criteria for patients for clinical trial subjects. The memory unit 16 stores, for example, a selection model. The memory unit 16 stores, for example, a graph generation model. The memory unit 16 stores, for example, an imaging diagnosis model. The memory unit 16 stores, for example, a prediction model. The selection model, the graph generation model, the imaging diagnosis model, and the prediction model may be stored in the storage means of the memory unit 16.

[0065] The terminal device 20 is, for example, a terminal device that accesses the clinical trial support device 10 and is used for processing to extract patients who are subject to a 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 a medical institution. An institution to which a hospital entrusts work related to a clinical trial is, for example, an SMO (Site Management Organization). A person in charge of an institution entrusted with work related to the clinical trial by a medical institution is, for example, a CRC (Clinical Research Coordinator). The terminal device 20 may also be a terminal device used by a person in charge of conducting a clinical trial at a pharmaceutical company or a person in charge of an institution entrusted with a clinical trial by a pharmaceutical company that conducts clinical trials of drugs. An institution entrusted with a clinical trial by a pharmaceutical company is, for example, a CRO (Contract Research Organization). A person in charge of an institution entrusted with a clinical trial by a pharmaceutical company is, for example, a CRA (Clinical Research Associate).

[0066] The terminal device 20 outputs the selection criteria to, for example, the acquisition unit 11 of the clinical trial support device 10. The terminal device 20 also acquires information about the patients who are the subject of the clinical trial from the output unit 15 of the clinical trial support device 10. The terminal device 20 then outputs the information about the patients who are the subject of the clinical trial to a display device (not shown).

[0067] The data management device 30 is a device that stores 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 information in the electronic medical record may be data entered by a nurse, a laboratory technician, a physical therapist, or a counselor. 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 listed in the electronic medical record. The data management device 30 outputs the data related to treatment to, for example, the acquisition unit 11 of the clinical trial support device 10.

[0068] The operation of the clinical trial support device 10 in the process of extracting patients who are subject to a clinical trial will be described below. Figure 8 shows an example of the flow of the process of the clinical trial support device 10 in extracting patients who are subject to a clinical trial.

[0069] The acquisition unit 11 acquires data related to the treatment of a patient (step S11).

[0070] When the data on the treatment of the patient is acquired, the complementing unit 12 complements missing data on the treatment from among the data used to select patients for clinical trial (step S12).

[0071] If data complementation has been completed for all target patients (Yes in step S13), the selection unit 13 selects patients to be included in the clinical trial based on the complemented data (step S14).

[0072] Once the patient to be a subject of the clinical trial is selected, the output unit 15 outputs information about the selected patient to be a subject of the clinical trial (step S15).

[0073] If there is a patient for whom data complementation has not been completed in step S13 (No in step S13), the process returns to step S12, and the complementing unit 12 complements missing data in the treatment-related data for the patient for whom data complementation has not been completed.

[0074] The clinical trial support device 10 complements missing data related to treatment among the data used to select patients for clinical trials. The clinical trial support device 10 selects patients for clinical trials based on the complemented data. The clinical trial support device 10 then outputs information about the selected patients for clinical trials. By selecting patients for clinical trials based on the complemented data in this way, for example, the number of patients who can be selected can be increased, allowing the clinical trial support device 10 to efficiently select patients for clinical trials.

[0075] Furthermore, by selecting patients for clinical trials so as not to cause bias in the selected patients for clinical trials, the clinical trial support device 10 can, for example, improve the effectiveness of clinical trial data in clinical trials conducted on the selected patients for clinical trials.

[0076] Furthermore, by selecting patients for the control group based on data related to past treatment, the clinical trial can be conducted by selecting only patients for the clinical trial group, so the clinical trial support device 10 can improve the efficiency of selecting patients for clinical trials. Furthermore, by selecting patients for the control group based on data related to past treatment, the administration of placebos can be suppressed, so the clinical trial support device 10 can improve the quality of medical care.

[0077] Furthermore, by selecting patients for the clinical trial based on the supplemented data, it is possible to select patients for the clinical trial from among patients with missing data, so the clinical trial support device 10 can select patients for the clinical trial even in clinical trials of pharmaceuticals for diseases and injuries with few cases. Furthermore, by selecting patients for the control group based on data related to past treatment, it is possible to select patients who are currently undergoing treatment as patients for the clinical trial group, so the clinical trial support device 10 can select patients for the clinical trial even in clinical trials of pharmaceuticals for diseases and injuries with few cases.

[0078] Furthermore, for example, by selecting patients for clinical trials based on the supplemented data, it is possible to select patients for clinical trials from among patients with missing data, and thus the clinical trial support device 10 can extract a wide range of patients for clinical trials. This allows the person in charge of selecting patients for clinical trials to select patients for clinical trials from a wide range of patients selected based on the supplemented data, for example, and make appropriate decisions regarding the selection of patients for clinical trials. Therefore, the clinical trial support device 10 can support decision-making regarding the selection of patients for clinical trials.

[0079] Each process in the clinical trial support device 10 may be distributed and executed among multiple information processing devices connected via a network. For example, the processes in the complementing unit 12 and the selecting unit 13 may be executed by different information processing devices. Which information processing device executes each process in the clinical trial support device 10 can be set as appropriate.

[0080] Each process in the clinical trial support device 10 can be realized by executing a computer program on a computer. Figure 9 shows an example of the configuration of a computer 100 that executes a computer program that performs each process in the clinical trial 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.

[0081] The CPU 101 reads and executes computer programs that perform each process from the storage device 103. The CPU 101 may be configured by a combination of multiple CPUs. The CPU 101 may also be configured by a combination of a CPU and another type of processor. For example, the CPU 101 may be configured by a combination of a CPU and a graphics processing unit (GPU). The memory 102 is configured by 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 by, 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 receives input from an operator and outputs display data, etc. The communication I / F 105 is an interface that transmits and receives data between the terminal device 20, the data management device 30, and other information processing devices. The terminal device 20 and the data management device 30 may also have the same configuration as the computer 100.

[0082] 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.

[0083] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0084] [Appendix 1] an acquisition means for acquiring data relating to the patient's treatment; A means for complementing missing data in the data related to the treatment among the data used to select patients to be treated in the clinical trial; a selection means for selecting patients to be included in the clinical trial based on the supplemented data; an output means for outputting information about the selected patient who is a subject of the clinical trial; A clinical trial support device equipped with:

[0085] [Appendix 2] the imputing means imputs missing data in the data related to the treatment using a graph showing relationships between patients generated based on the data related to the treatment of the patients; 1. A clinical trial support device as described in Appendix 1.

[0086] [Appendix 3] the complementing means preserves the missing data using unstructured data from the data related to the treatment; 1. A clinical trial support device as described in Appendix 1.

[0087] [Appendix 4] the complementing means generates a feature vector for each patient as data obtained by complementing the missing data, based on a graph showing relationships between patients generated based on data related to the treatment of the patients and on selection criteria for patients to be clinical trial subjects; 1. A clinical trial support device as described in Appendix 1.

[0088] [Appendix 5] The imputation means imputs missing data in the data related to the treatment among the data used to calculate the degree of conformance with the selection criteria for patients to be included in the clinical trial. A clinical trial support device according to any one of appendices 1 to 3.

[0089] [Appendix 6] The selection means selects patients to be treated in the clinical trial from patients who met the selection criteria at a past point in time and for whom data on treatment covering the period of the clinical trial has been recorded since the time the patients met the selection criteria. 6. A clinical trial support device as described in Appendix 5.

[0090] [Appendix 7] The selection means selects the patient to be a subject of the clinical trial further based on data regarding the treatment of the patient already selected as the patient to be a subject of the clinical trial. A clinical trial support device according to any one of appendices 1 to 3.

[0091] [Appendix 8] and further comprising a prediction means for predicting data related to the treatment at a predetermined time point based on the data related to the treatment. A clinical trial support device according to any one of appendices 1 to 3.

[0092] [Appendix 9] The selection means selects at least a portion of the patients in the control group based on data of the patients in the clinical trial group who have already been selected as patients to be tested in the clinical trial. 8. A clinical trial support device as described in Appendix 7.

[0093] [Appendix 10] The selection means further selects patients in the clinical trial group based on data of patients in the clinical trial group already selected as patients to be included in the clinical trial. 8. A clinical trial support device as described in Appendix 7.

[0094] [Appendix 11] The selection means selects patients to be the subjects of the clinical trial based on the feature vectors of each patient. 5. A clinical trial support device as described in Appendix 4.

[0095] [Appendix 12] the imputation means imputing missing data in the data related to the treatment predicted by the prediction means; 9. A clinical trial support device as described in Appendix 8.

[0096] [Appendix 13] The prediction means predicts data regarding treatment at the predetermined time point for the patient who is the subject of the clinical trial selected by the selection means. 9. A clinical trial support device as described in Appendix 8.

[0097] [Appendix 14] The predetermined time point is the start of the clinical trial or a time point during the clinical trial; 9. A clinical trial support device as described in Appendix 8.

[0098] [Appendix 15] Obtain data on patient treatment, Among the data used to select patients for clinical trials, data missing in the data related to the treatment are supplemented; Based on the supplemented data, patients will be selected for the clinical trial. outputting information about the selected patients who are the subjects of the clinical trial; Clinical trial support methods.

[0099] [Appendix 16] A process of obtaining data relating to the patient's treatment; A process of complementing missing data in the data related to the treatment among the data used to select patients to be clinical trial subjects; A process of selecting patients for clinical trials based on the supplemented data; A process of outputting information about the selected clinical trial subject patient; A clinical trial support program that runs the above on a computer.

[0100] Furthermore, some or all of the configurations described in Supplementary Notes 2 to 14, which are dependent on Supplementary Note 1, may also be dependent on Supplementary Notes 15 and 16 in the same dependent relationship as Supplementary Notes 2 to 14. Furthermore, not limited to Supplementary Notes 1, 15, and 16, some or all of the configurations described as Supplements may be made dependent on various hardware, software, various recording means for recording software, or systems, within the scope of each of the above-mentioned embodiments.

[0101] 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. [Explanation of symbols]

[0102] 10 Clinical trial support equipment 11 Acquisition Department 12 Complementary Section 13 Selection Department 14 Prediction Department 15 Output section 16 Memory section 20 Terminal equipment 30 Data management device 100 computers 101 CPU 102 memory 103 Storage device 104 Input / Output Interface 105 Communication I / F

Claims

1. an acquisition means for acquiring data relating to the patient's treatment; A means for complementing missing data in the data related to the treatment among the data used to select patients to be clinical trial subjects; a selection means for selecting patients to be included in the clinical trial based on the supplemented data; an output means for outputting information about the selected patient who is a subject of the clinical trial; A clinical trial support device equipped with:

2. the imputing means imputs missing data in the data related to the treatment using a graph showing relationships between patients generated based on the data related to the treatment of the patients; The clinical trial support device according to claim 1.

3. the complementing means preserves the missing data using unstructured data from the data related to the treatment; The clinical trial support device according to claim 1.

4. the complementing means generates a feature vector for each patient as data obtained by complementing the missing data, based on a graph showing relationships between patients generated based on data related to the treatment of the patients and on selection criteria for patients to be clinical trial subjects; The clinical trial support device according to claim 1.

5. The imputation means imputs missing data in the data related to the treatment among the data used to calculate the degree of conformance with the selection criteria for patients to be included in the clinical trial.

4. A clinical trial support device according to claim 1.

6. The selection means selects patients to be treated in the clinical trial from patients who met the selection criteria at a past point in time and for whom data on treatment covering the period of the clinical trial has been recorded since the time the patients met the selection criteria. The clinical trial support device according to claim 5.

7. The selection means selects the patient to be a subject of the clinical trial further based on data regarding the treatment of the patient already selected as the patient to be a subject of the clinical trial.

4. A clinical trial support device according to claim 1.

8. and further comprising a prediction means for predicting data related to the treatment at a predetermined time point based on the data related to the treatment.

4. A clinical trial support device according to claim 1.

9. Obtain data on patient treatment, Among the data used to select patients for clinical trials, data missing in the data related to the treatment are supplemented; Based on the supplemented data, patients will be selected for the clinical trial. outputting information about the selected patients who are the subjects of the clinical trial; Clinical trial support methods.

10. A process of obtaining data relating to the patient's treatment; A process of complementing missing data in the data related to the treatment among the data used to select patients to be clinical trial subjects; A process of selecting patients for clinical trials based on the supplemented data; A process of outputting information about the selected clinical trial subject patient; A clinical trial support program that runs the above on a computer.

Images