Data generation device, data generation method, and data generation program

Abstract

The present invention provides a data generation device that can easily generate time-series data related to a patient's medical condition. [Solution] The data generation device 10 comprises an acquisition unit 11, a virtual patient data generation unit 12, a time-series data generation unit 14, and an output unit 16. The acquisition unit 11 acquires data related to the patient's treatment. The virtual patient data generation unit 12 generates treatment data for each stage of the virtual patient's illness based on the patient's treatment data. The time-series data generation unit 14 generates time-series data related to the virtual patient's illness from the generated virtual patient treatment data. The output unit 16 outputs the generated time-series data related to the virtual patient's illness. With this configuration, the data generation device 10 can support decision-making regarding the patient's illness.

Description

Technical Field

[0001] This disclosure relates to a data generation device and the like.

Background Art

[0002] For formulating a clinical trial plan in a pharmaceutical company, time-series data regarding a patient's medical condition may be used. Also, time-series data regarding a patient's medical condition may be used for treatment in a medical institution. Such time-series data regarding a patient's state is also called a patient journey. For example, by referring to time-series data regarding a patient's medical condition for each case, it may be possible to formulate a more accurate plan. Also, data used for formulating a plan regarding a patient may be generated using an information processing system.

[0003] The information processing system of Patent Document 1 identifies a treatment policy based on guideline data corresponding to a patient's disease or case. Then, the information processing system of Patent Document 1 displays the identified treatment policy.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the technique described in Patent Document 1, it may be difficult to generate time-series data regarding a patient's medical condition.

[0006] An object of this disclosure is to provide a data generation device and the like that can easily generate time-series data regarding a patient's medical condition in order to solve the above problems.

Means for Solving the Problems

[0007] To solve the above problems, the data generation device of this disclosure comprises: acquisition means for acquiring data relating to patient treatment; virtual patient data generation means for generating data relating to treatment at each stage of a virtual patient's condition based on the patient treatment data; time series data generation means for generating time series data relating to the virtual patient's condition from the generated virtual patient treatment data; and output means for outputting the generated time series data relating to the virtual patient's condition.

[0008] The data generation method disclosed herein acquires data related to the patient's treatment, generates data related to the treatment at each stage of the virtual patient's condition based on the patient's treatment data, generates time-series data related to the virtual patient's condition from the generated treatment data, and outputs the generated time-series data related to the virtual patient's condition.

[0009] The data generation program disclosed herein is a treatment support program that causes a computer to perform the following processes: acquiring data related to a patient's treatment; generating data related to treatment at each stage of a virtual patient's condition based on the patient's treatment data; generating time-series data related to the virtual patient's condition from the generated treatment data; and outputting the generated time-series data related to the virtual patient's condition. [Effects of the Invention]

[0010] According to this disclosure, time-series data regarding the patient's medical condition can be easily generated. [Brief explanation of the drawing]

[0011] [Figure 1] This figure shows an example of the configuration of the data generation system described in this disclosure. [Figure 2] This figure shows an example of the configuration of the data generation device in this disclosure. [Figure 3] This diagram schematically illustrates an example of the process for generating data related to the treatment of a virtual patient in this disclosure. [Figure 4] This diagram schematically illustrates an example of the process for generating data related to the treatment of a virtual patient in this disclosure. [Figure 5] This figure shows an example of a display screen for time-series data related to the patient's condition as described in this disclosure. [Figure 6] This figure shows an example of a display screen for time-series data related to the patient's condition as described in this disclosure. [Figure 7] This figure shows an example of a display screen for time-series data related to the patient's condition as described in this disclosure. [Figure 8] This figure shows an example of the operation flow of the data generation device in this disclosure. [Figure 9] This figure shows an example of the hardware configuration of the data generation device in this disclosure. [Modes for carrying out the invention]

[0012] Embodiments of this disclosure will be described in detail with reference to the figures. Figure 1 is a diagram showing an example of the configuration of a data generation system. The data generation system comprises a data generation device 10, a terminal device 20, and a data management device 30. The data generation device 10 is connected to the terminal device 20, for example, via a network. The data generation device 10 is also 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 can be set as appropriate.

[0013] The data generation system generates, for example, time-series data on a patient's medical condition. For example, the data generation system generates time-series data on a virtual patient's medical condition. A virtual patient is a hypothetical person who is assumed to be suffering from the disease for which time-series data on medical condition is generated. Data on the virtual patient's treatment is also data generated by the data generation system. A virtual patient is a person who, if a specialist were to review the treatment data generated by the data generation system, is highly likely to be diagnosed with the disease in question.

[0014] The time-series data regarding the condition of the virtual patient is, for example, data regarding the condition of the patient at each stage in the time series from the discovery of the disease to the completion of treatment. The time-series data regarding the condition of the virtual patient may also be data regarding the condition of the patient at each stage in the time series from hospitalization to discharge. Further, the discovery of the disease may include, for example, the patient's recognition of a physical abnormality. Further, the completion of treatment may include, for example, the end of treatment at the end stage or the death of the patient.

[0015] For example, when a virtual patient is discovered to have lung cancer at stage 0 and dies after progressing to stage 3, the time-series data regarding the condition of the virtual patient is data regarding the condition of the patient at each stage from stage 0 to 3. The time-series data regarding the condition of the patient may also be the number of patients at each stage of the condition.

[0016] The time-series data regarding the condition of the virtual patient is, for example, data indicating at least one of the patient's state and treatment details at each stage of the condition. The patient's state is, for example, data of one or more items among test results, medical staff's findings, the patient's chief complaint, and the patient's emotions. The medical staff are, for example, doctors, nurses, pharmacists, medical laboratory technicians, clinical therapists, psychotherapists, caregivers, counselors, or administrative staff of a medical institution. The medical staff are not limited to the above. Further, the patient's state is not limited to the above. Further, the treatment details are, for example, data of one or more items among the diagnosis, tests, medication, rehabilitation, counseling, physical care, ward, equipment to be installed, and diet that the medical staff perform on the patient. The treatment details are not limited to the above. Further, the time-series data regarding the condition of the patient may also be the medical expenses at each stage of the condition. The time-series data regarding the condition of the patient is not limited to the above.

[0017] The stage of the medical condition is, for example, the stage of the progression of the medical condition or the progression of the treatment. For example, the stage of the medical condition is the stage of the progression of the medical condition or the progression of the treatment. For example, in the case of a disease where the stage of the medical condition is the progression of the medical condition and the degree of progression of the medical condition is represented using stages, the stage of the medical condition is the stage. Also, when the stage of the medical condition is the stage of the progression of the treatment, the stage of the medical condition is, for example, the stages of each of the visit, hospitalization, surgery, rehabilitation, and discharge. Also, when the stage of the medical condition is the stage of the progression of the treatment, the stage of the medical condition may be the stages of the recognition of physical abnormalities by the patient, information collection, visit, diagnosis, treatment, and support respectively. The stage of the medical condition is not limited to the above.

[0018] The data generation system generates, for example, data regarding the treatment of a virtual patient at each stage of the medical condition based on data regarding the treatment of a patient. Then, the data generation system generates time-series data regarding the patient's medical condition based on the generated data regarding the treatment of the virtual patient. The data regarding the treatment of the virtual patient is, for example, virtual data generated by the data generation system. For example, the data regarding the treatment of the virtual patient is data that would be described in an electronic medical record as a record of treatment if the virtual patient actually existed. Specific examples of the data regarding the treatment of the virtual patient will be described later.

[0019] The data generation system generates, for example, data regarding the treatment of a virtual patient using a generation model. The generation model is, for example, a machine learning model that takes data regarding the treatment of a patient as input and generates data regarding the treatment of a virtual patient. Specific examples of the generation model will be described later. The data generation system can generate time-series data regarding the patient's medical condition even in the case of a rare disease by generating time-series data regarding the medical condition of the virtual patient based on the generated data regarding the treatment of the virtual patient.

[0020] Here, an example of the configuration of the data generation device 10 will be described. Figure 2 shows an example of the configuration of the data generation device 10. The data generation device 10 basically comprises an acquisition unit 11, a virtual patient data generation unit 12, a time-series data generation unit 14, and an output unit 16. The data generation device 10 also further comprises, for example, a selection unit 13, a prediction unit 15, and a storage unit 17.

[0021] The acquisition unit 11 acquires data related to the patient's treatment. The patient's treatment data is used, for example, to generate data related to the treatment of a virtual patient. The patient's treatment data is, for example, patient treatment data at a medical institution. The patient's treatment data is, for example, a record of medical procedures performed on the patient. The patient's treatment data is, for example, a record of one or more items from diagnosis, examination, medication, surgery, follow-up, and the patient's condition. The patient's condition includes, for example, information on one or more items from disease information, complication information, biomarkers, disease status, guideline score, treatment effect, and test results. 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, imaging tests, and genomic tests. The test results are not limited to the above. The patient's treatment data may also include information on the person who performed the medical procedure. Information on the person who performed the medical procedure is, for example, information indicating the medical professional who performed the medical procedure on the patient. Information identifying healthcare professionals who performed medical procedures on a patient includes, for example, the names or identifiers of doctors, nurses, pharmacists, and physical therapists.

[0022] The acquisition unit 11 acquires data from the electronic medical record, for example, as data related to the patient's treatment. The data related to the patient's treatment may also be examination data. If the data related to the patient's treatment is examination data, it may also be image data for diagnostic imaging. Alternatively, the data related to the patient's treatment may also be data recorded on the medical claim form. The data related to treatment may also include, for example, the patient's perception of their medical condition.

[0023] The acquisition unit 11 may acquire data regarding the patient's health condition before visiting a hospital as data regarding the patient's treatment. For example, the acquisition unit 11 may acquire data regarding the patient's treatment when the patient visited another hospital as data regarding the patient's health condition before visiting a hospital. Alternatively, the acquisition unit 11 may acquire information indicating what the patient or people around the patient perceived about the patient's physical condition as data regarding the patient's treatment. For example, if the patient was experiencing pain in a specific body part, the acquisition unit 11 may acquire information indicating the body part where the pain occurred, the duration of the pain, and the severity of the pain as data regarding the patient's treatment.

[0024] The acquisition unit 11 may further acquire data on the patient's health status after discharge as data on the patient's treatment. For example, the acquisition unit 11 may acquire data on the patient's health status at home or in a care facility after discharge as data on the patient's treatment. Examples of care facilities include nursing homes, long-term care medical facilities, or special nursing homes for the elderly. The care facilities are not limited to those mentioned above. The acquisition unit 11 may also acquire data on the patient's treatment at the medical institution to which the patient is transferred. The data on the patient's treatment is not limited to those mentioned above.

[0025] The acquisition unit 11 may acquire data related to the treatment of the patient whose future condition is to be predicted. For example, the acquisition unit 11 acquires data related to the patient's treatment up to the point in time when the patient performs the process related to predicting the future condition. The acquisition unit 11 may also acquire data related to the patient's treatment from, for example, the data management device 30. The acquisition unit 11 may also acquire data related to the patient's treatment from the terminal device 20.

[0026] The acquisition unit 11 acquires, for example, information specifying the target disease. The target disease is, for example, a disease for which time-series data on the symptoms of a hypothetical disease is to be generated. For example, when time-series data is used to understand the symptoms of clinical trial patients, the disease for which the time-series data is used is the disease for which the drug being tested is used. The acquisition unit 11 acquires, for example, information specifying the target disease from the terminal device 20.

[0027] The virtual patient data generation unit 12 generates treatment data for each stage of a virtual patient's illness based on data related to the patient's treatment. For example, the virtual patient data generation unit 12 generates treatment data for each stage of a virtual patient's illness for a virtual patient suffering from a target disease. The virtual patient data generation unit 12 generates treatment data for each stage of a virtual patient's illness using, for example, a generative model. The generative model is, for example, a machine learning model that generates treatment data for each stage of an illness from data related to the patient's treatment. The virtual patient data generation unit 12 generates treatment data for multiple virtual patients using, for example, a generative model.

[0028] The virtual patient data generation unit 12 generates data related to the treatment of virtual patients, for example, using a language model as the generation model. For example, a large-scale language model 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. Alternatively, Claude3, Claude3.5, 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) may be used as the language model. The language model used to generate data related to the treatment of virtual patients is not limited to those listed above.

[0029] The virtual patient data generation unit 12 generates data related to the treatment of a virtual patient, for example, by taking a prompt that instructs the generation of data related to the treatment of a virtual patient as input to a language model. The prompt that instructs the generation of data related to the treatment of a virtual patient may include, for example, an instruction to generate data related to the treatment of the virtual patient for each stage of the disease. The prompt may also include, for example, the name of the disease that the virtual patient to be generated is suffering from. The name of the disease that the virtual patient to be generated is, for example, information that specifies the target disease. The prompt may also include attributes of the virtual patient to be generated.

[0030] When generating an electronic medical record as data related to the treatment of a virtual patient, the virtual patient data generation unit 12 generates data for the virtual patient's electronic medical record as data related to the virtual patient's treatment, for example, by taking a prompt containing information indicating the generation of an electronic medical record for a virtual patient as input to the language model. For example, the virtual patient data generation unit 12 generates data related to the treatment of a virtual patient by taking the prompt "Create an electronic medical record for a virtual patient suffering from stage 3 liver cancer" as input to the language model. The information indicating the generation of an electronic medical record for a virtual patient may include one or more items from among the virtual patient's attributes, the disease it is suffering from, the number of virtual patients, and the stage of the disease condition to be generated.

[0031] The virtual patient data generation unit 12 may generate treatment data for each stage of the disease for a single virtual patient. For example, the virtual patient data generation unit 12 generates treatment data for a virtual patient suffering from stage 3 liver cancer for stages 0, 1, 2, and 3. For example, the virtual patient data generation unit 12 takes the prompt "Create electronic medical record data for stage 0 of a virtual patient suffering from stage 3 liver cancer" as input to the language model and generates electronic medical record data for stage 0. The virtual patient data generation unit 12 also generates electronic medical record data for stages 1, 2, and 3, respectively. Then, the virtual patient data generation unit 12 uses the electronic medical record data for each stage from stage 0 to stage 3 to generate an electronic medical record that includes descriptions for each stage from stage 0 to stage 3.

[0032] The virtual patient data generation unit 12 generates treatment data for a predetermined number of people, for example. The predetermined number is set to be sufficient for understanding the progression of the disease for each target disease. The predetermined number may also be set to be sufficient for understanding the progression of the disease for each attribute of the virtual patient. The predetermined number may also be set to be sufficient for understanding the progression of the disease for each stage of the disease that the patient will eventually reach. Furthermore, the predetermined number may be set to be sufficient for understanding the progression of the disease for each stage of the disease at which the patient is confirmed to have the disease.

[0033] When generating data related to the treatment of a virtual patient using a language model, the virtual patient data generation unit 12 may use data related to the patient's treatment as a data source and generate data related to the treatment of the virtual patient using RAG (Retrieval Augmented Generation) technology. For example, the virtual patient data generation unit 12 may use the patient's electronic medical record as a data source for data related to the patient's treatment and generate data from the virtual patient's electronic medical record as data related to the treatment of the virtual patient.

[0034] When generating an electronic medical record for a virtual patient suffering from a rare disease, the virtual patient data generation unit 12 generates the electronic medical record for the virtual patient using, for example, data on the treatment of patients actually suffering from the rare disease as a data source. A rare disease is, for example, a disease in which few patients develop the condition. In other words, a rare disease is, for example, a disease for which there is insufficient data on the treatment of patients. When generating an electronic medical record for a virtual patient suffering from a rare disease, the virtual patient data generation unit 12 may further use, for example, data on the treatment of patients actually suffering from a disease similar to the rare disease as a data source to generate the electronic medical record for the virtual patient. Similar diseases mean, for example, that at least the type of disease, causative factors, and site of occurrence are the same.

[0035] Figure 3 schematically illustrates an example of a process for generating data related to the treatment of virtual patients. In the example in Figure 3, electronic medical records for n virtual patients suffering from disease A are generated using the electronic medical records of patients P1, P2, and P3, who are suffering from disease A, as data sources. The virtual patient data generation unit 12 generates the electronic medical records of virtual patients by, for example, using a prompt containing instructions to generate the electronic medical records of virtual patients as input to a language model that uses this prompt as a generation model.

[0036] The virtual patient data generation unit 12 may generate data regarding the virtual patient's complications as data regarding the virtual patient's treatment. For example, the virtual patient data generation unit 12 generates electronic medical record data for virtual patients who have developed complications, using prompts that include the names of each disease included in the complications as information indicating that an electronic medical record for the virtual patient is to be generated.

[0037] Figure 4 schematically illustrates an example of a process for generating data related to the treatment of a virtual patient suffering from complications. The example in Figure 4 shows a process for generating data related to the treatment of a virtual patient suffering from both disease A and disease B. In the example in Figure 4, electronic medical records for n virtual patients suffering from both diseases A and B are generated using the electronic medical records of patients Pa1, Pa2, and Pa3 suffering from disease A, and the electronic medical records of m patients suffering from disease B, as data sources. The virtual patient data generation unit 12 generates the electronic medical records of virtual patients, for example, by using a language model that uses a prompt containing instructions to generate electronic medical records for virtual patients suffering from both diseases A and B as input to the generation model.

[0038] The virtual patient data generation unit 12 may generate data on treatment at each stage of the virtual patient's condition, including, for example, before visiting a hospital. The virtual patient data generation unit 12 generates data on the virtual patient's treatment using, for example, data indicating the patient's health status before visiting a hospital as a data source. The data indicating the patient's health status before visiting a hospital may be, for example, data from a health checkup. The data indicating the patient's health status before visiting a hospital may include at least one of the patient's perception of their own physical condition and the perception of the patient's physical condition by people around them. The data indicating the patient's health status before visiting a hospital may also be, for example, data on treatment received at other medical institutions before visiting a hospital.

[0039] Furthermore, the virtual patient data generation unit 12 may generate data related to treatment at each stage of the virtual patient's condition, including after discharge. For example, the virtual patient data generation unit 12 generates data related to the virtual patient's treatment using data indicating the patient's health status after discharge as a data source. Data indicating the patient's health status after discharge may be, for example, health checkup data. Data indicating the patient's health status after discharge may also be, for example, the results of outpatient examinations. Data indicating the patient's health status after discharge may include at least one of the patient's perception of their own physical condition and the perception of the patient's physical condition by people around them. In addition, data indicating the patient's health status after discharge may also be, for example, data related to treatment at other medical institutions.

[0040] The virtual patient data generation unit 12 may generate data on the treatment of a virtual patient using statistical data on treatment. For example, the virtual patient data generation unit 12 generates data on the treatment of a virtual patient by replacing data for a predetermined item from the treatment data so that the values ​​of the data for that predetermined item are included in the statistical data. For example, the virtual patient data generation unit 12 generates data on the treatment of a virtual patient by replacing data for a predetermined item based on the mean, mode, or median of the patient's treatment data and the variance. For example, statistical data on the treatment of patients suffering from diseases similar to the target disease may be used for the statistical data on the treatment of patients suffering from the target disease.

[0041] The specified items are, for example, set for each target disease. The specified items are, for example, items from the treatment data that are greatly affected by the disease. The items are, for example, the type of data. For example, if the data is the measurement results of blood glucose levels and uric acid levels, then blood glucose levels and uric acid levels would each be items. The specified items may also be items from the treatment data that require attention when treating the disease. Alternatively, the specified items may be items from the treatment data that require attention in treatment but have a low relationship with the disease.

[0042] Furthermore, the virtual patient data generation unit 12 may generate data related to the treatment of a virtual patient using an image generation model that generates image data for diagnostic imaging as a generation model. For example, the virtual patient data generation unit 12 uses an image generation model to generate image data for diagnostic imaging at each stage of the disease. Then, the virtual patient data generation unit 12 generates data related to the treatment of a virtual patient by determining the disease state based on the image data for diagnostic imaging at each stage of the disease. For example, if the image data for diagnostic imaging is data from a CT (computed tomography) examination for tumor examination, the image generation model generates multiple image data of the same tumor location but with different sizes of tumors. For example, the image generation model generates multiple image data of a growing tumor. The image generation model may also generate multiple image data of a shrinking tumor as a result of treatment. The image generation model may also generate images of a new tumor that has developed at a different location from an already existing tumor. Then, the virtual patient data generation unit 12 generates data related to treatment at each stage of the disease based on the size of the tumors shown in the images generated by the image generation model.

[0043] The image generation model is generated, for example, by machine learning using GANs (Generative Adversarial Networks). The algorithm for generating the image generation model is not limited to the above. The image generation model is generated, for example, by an external device to the data generation device 10. Furthermore, the generation model may be a machine learning model capable of processing both language data and image data.

[0044] Data relating to the treatment of the virtual patient may be generated by an external information processing device of the data generation device 10. In this case, the virtual patient data generation unit 12 outputs a request to an external information processing device on which the generation model operates to generate data relating to the treatment of the virtual patient. The virtual patient data generation unit 12 then obtains the data relating to the treatment of the virtual patient from the external information processing device on which the generation model operates.

[0045] When the process of generating data related to the treatment of a virtual patient is performed in a device outside the data generation device 10, the virtual patient data generation unit 12 outputs a prompt to an information processing device on which a language model that performs the process of generating data related to the treatment of a virtual patient based on a prompt is operating, for example, a prompt containing information indicating that an electronic medical record for the virtual patient is to be generated. The virtual patient data generation unit 12 then retrieves the generated data related to the treatment of the virtual patient from the information processing device that generated the data related to the treatment of the virtual patient.

[0046] The selection unit 13 selects, for example, data from the data on the treatment of virtual patients generated by the virtual patient data generation unit 12 that is suitable as data for the target disease. The selection unit 13 selects data on the treatment of virtual patients to be used for extracting time-series data on the condition of virtual patients, for example, based on the suitability of the data on the treatment of virtual patients generated by the virtual patient data generation unit 12. The suitability is, for example, an index that indicates the suitability of the data for use in extracting time-series data on the condition of virtual patients.

[0047] The selection unit 13 calculates, for example, the suitability of the data related to the treatment of virtual patients generated by the virtual patient data generation unit 12. The selection unit 13 then selects the data related to the treatment of virtual patients whose calculated suitability meets predetermined criteria as the data related to the treatment of virtual patients to be used for extracting time-series data related to the condition of virtual patients. The predetermined criteria are set, for example, so that when the suitability meets the criteria, the data related to the treatment of virtual patients that meet the criteria becomes appropriate data as patient data for the target disease.

[0048] For suitability, for example, the similarity between reference data and data on the treatment of the generated virtual patient can be used. Similarity is an index that shows the degree of agreement between the reference data for each disease and the data on the treatment of the virtual patient. The reference data for each disease is set based on, for example, assumed values ​​for treatment data when a patient has that disease. Also, if the data on the treatment of the virtual patient is an electronic medical record, the similarity may be the similarity between the electronic medical record of a patient who actually has the target disease and the electronic medical record of the virtual patient.

[0049] The selection unit 13 calculates the similarity of data related to the treatment of a virtual patient for the target disease, for example, using a similarity calculation model. The similarity calculation model is a machine learning model that calculates the similarity of data related to the treatment of a virtual patient for the target disease from data related to the treatment of the virtual patient. The similarity calculation model converts, for example, the reference data of the target disease and the data of items included in the reference data from the data related to the treatment of the virtual patient into feature vectors. Then, the selection unit 13 calculates the similarity between the converted feature vectors, for example, Euclidean distance or cosine similarity is used to calculate the similarity between feature vectors. The similarity calculation model is generated, for example, by a device outside the data generation device 10.

[0050] The suitability score may be an index based on a score set for each treatment-related data. The selection unit 13 calculates a score for each treatment-related data of a virtual patient based on the score criteria set for each treatment-related data. The selection unit 13 calculates the suitability score for each treatment-related data of a virtual patient based on the sum of the scores. The selection unit 13 then selects the treatment-related data of virtual patients whose calculated suitability scores meet predetermined criteria as the treatment-related data of virtual patients to be used for extracting time-series data on the condition of the virtual patients. Furthermore, the method for calculating the suitability score is not limited to the above.

[0051] The selection unit 13 may select data relating to the treatment of virtual patients for the target disease using a selection model. The selection model is, for example, a machine learning model that determines whether data relating to the treatment of virtual patients is suitable as data relating to the treatment of patients with the target disease. The selection unit 13 selects, for example, the data relating to the treatment of virtual patients that the selection model has determined to be suitable as data relating to the treatment of patients with the target disease. The selection model is generated by learning the relationship between the selection criteria and the data relating to the treatment of virtual patients and whether or not it is suitable as data relating to the treatment of patients with the target disease. The selection model is generated, for example, by deep learning using a neural network. The machine learning algorithm used to generate the selection model is not limited to the above. The selection model is generated, for example, by a device outside the data generation device 10.

[0052] The time-series data generation unit 14 generates time-series data relating to the virtual patient's condition from the data relating to the treatment of the generated virtual patient. The time-series data relating to the virtual patient's condition is, for example, time-series data relating to at least one of the treatments performed on the patient and the patient's condition. The time-series data generation unit 14 generates data in the form of a patient journey as time-series data relating to the virtual patient's condition. The time-series data generation unit 14 extracts data for each stage of the condition from the data relating to the virtual patient's treatment and generates time-series data relating to the virtual patient's condition.

[0053] For example, if the disease stage is defined by stages, the time-series data generation unit 14 extracts data on the disease stage for each stage from the data on the treatment of the virtual patient and generates data in the order of the stages. Also, for example, if the disease stages are classified into consultation, hospitalization, surgery, rehabilitation, and discharge, and each stage is in the above order in the time series, the time-series data generation unit 14 extracts data on the disease stage for each of the above-mentioned stages from the data on the treatment of the virtual patient and generates data in the order of the time series as time-series data on the disease stage of the virtual patient.

[0054] The time-series data generation unit 14 generates, for example, time-series data relating to the disease state, including the treatment content at each stage of the disease state. The time-series data generation unit 14 generates, for example, time-series data relating to the disease state, including information showing the procedures that each healthcare professional performs on the patient at each stage of the disease state. For example, if the healthcare professional is a doctor, the time-series data generation unit 14 generates, for example, information showing the treatment actions, medications, and tests that the doctor performs on the patient at each stage of the disease state. For example, if the healthcare professional is a nurse, the time-series data generation unit 14 generates, for example, information showing the tests, checks, and advice that the nurse performs on the patient at each stage of the disease state. The time-series data generation unit 14 may also generate time-series data relating to the disease state, including the findings of healthcare professionals at each stage of the disease state. For example, the time-series data generation unit 14 generates time-series data relating to the disease state, including the findings of a doctor at each stage of the disease state.

[0055] The time-series data generation unit 14 generates, for example, the physical state at each stage of the illness as time-series data related to the illness. The time-series data generation unit 14 generates, for example, information showing the doctor's findings, the treatments being performed, the content of medications prescribed, and the test results at each stage of the illness as time-series data related to the illness. The time-series data generation unit 14 may also generate the patient's chief complaint at each stage of the illness as time-series data related to the illness. The patient's chief complaint may include the patient's feelings at each stage of the illness.

[0056] The time-series data generation unit 14 may generate the number of patients at each stage of the disease as time-series data related to the disease. For example, the time-series data generation unit 14 extracts the disease stages for each time progression from time-series data related to the treatment of each of several virtual patients. Then, for example, the time-series data generation unit 14 generates information showing the ratio of people at each stage of the disease as time-series data related to the disease. The ratio of people at each stage of the disease is, for example, the number of virtual patients corresponding to each stage of the disease relative to the total number of virtual patients suffering from the target disease at a certain point in the time series. The time-series data generation unit 14 may generate the number of people at each stage of the disease at each point in the time series as time-series data related to the disease, based on the number of people suffering from the target disease at the starting point.

[0057] For example, if the target disease is lung cancer, the time-series data generation unit 14 generates the number of patients for each stage as time-series data related to the disease state. For example, the time-series data generation unit 14 extracts the stage for each period of time from the time-series data related to the treatment of multiple virtual patients suffering from lung cancer. Then, for example, the time-series data generation unit 14 generates information indicating the number of virtual patients corresponding to each stage as time-series data related to the disease state for each period of time. The time-series data generation unit 14 may also generate information indicating the ratio of the number of virtual patients corresponding to each stage as time-series data related to the disease state for each period of time.

[0058] The prediction unit 15 predicts, for example, the trend of data related to the treatment of a target patient. The target patient is, for example, a patient who is to be treated by referring to time-series data on the medical condition of a virtual patient. The target patient may also be a patient who is the subject of a clinical trial when a clinical trial is conducted by referring to time-series data on the medical condition of a virtual patient. The prediction unit 15 extracts, for example, virtual patients whose data on treatment in the initial stages is similar to that of the target patient. The initial stages are, for example, a predetermined period from the start of treatment. The predetermined period is, for example, the period from the start of treatment until a full treatment plan is decided. The decision on a full treatment plan is, for example, the decision on whether or not to perform surgery and the decision on the treatment plan after surgery. The predetermined period is not limited to the above. The prediction unit 15 extracts virtual patients whose data on treatment in the initial stages is similar to that of the target patient, based on the similarity between the data on treatment in the initial stages of the target patient and the generated data on treatment of virtual patients.

[0059] The prediction unit 15 predicts the trend of data related to the treatment of the target patient, for example, based on time-series data related to the medical condition of the extracted virtual patients. The prediction unit 15 predicts the trend of data related to the treatment of the target patient, for example, by assuming that the data related to the treatment of the target patient will trend in a similar manner to the time-series data related to the medical condition of the extracted virtual patients. Furthermore, if the data related to the treatment of the target patient in the initial stages is similar to that of multiple virtual patients, the prediction unit 15 may predict the trend of data related to the treatment of the target patient for multiple patterns based on the time-series data related to the medical condition of each of the similar virtual patients.

[0060] The prediction unit 15 calculates the similarity between data on the target patient's initial treatment and data on the generated virtual patient's treatment, for example, using a calculation model. The calculation model is a machine learning model that calculates the similarity between data on the target patient's initial treatment and data on the generated virtual patient's treatment. The calculation model converts the data on the target patient's initial treatment and the data on the generated virtual patient's treatment into feature vectors. The prediction unit 15 then calculates the similarity between the converted feature vectors, for example, using Euclidean distance or cosine similarity. The calculation model is generated, for example, by a device outside the data generation device 10.

[0061] The prediction unit 15 may use a prediction model to predict data regarding the treatment of the target patient in stages later than the initial stage, based on data regarding the initial stage of the target patient's treatment. The prediction model is, for example, a machine learning model that predicts data regarding treatment in stages later than the initial stage in a time series from the patient's initial stage data. 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 is also generated, for example, by a device outside the data generation device 10. The prediction unit 15 may also extract data regarding the treatment of a virtual patient that is similar to the prediction result from the prediction model. For example, the prediction unit 15 extracts data regarding the treatment of a virtual patient that is similar to the prediction result from the prediction model in the same way as when extracting a virtual patient whose initial stage of treatment data is similar to that of the target patient.

[0062] The output unit 16 outputs time-series data relating to the medical condition of the generated virtual patients. For example, the output unit 16 outputs at least one of the patient's condition and treatment content for each stage of the medical condition as time-series data relating to the medical condition of the virtual patients. The output unit 16 may also output a graph of the time-series data relating to the medical condition of each virtual patient. For example, the output unit 16 outputs time-series data relating to the medical condition of each virtual patient using a graph with the stage of the medical condition on the vertical axis and the passage of time on the horizontal axis. The output unit 16 may also output the number of virtual patients for each stage of the medical condition as time-series data relating to the medical condition of the virtual patients.

[0063] The output unit 16 may output prediction results for data regarding the treatment of the target patient at a stage later than the initial stage. The output unit 16 may also output data regarding the treatment of a virtual patient similar to the prediction result. For example, the output unit 16 may output data regarding the treatment of a virtual patient whose initial treatment data is similar to that of the target patient. Furthermore, the output unit 16 may output data regarding the treatment of multiple virtual patients whose initial treatment data is similar to that of the target patient.

[0064] The output unit 16 outputs time-series data relating to the medical condition of the generated virtual patient to, for example, the terminal device 20. The output unit 16 also outputs time-series data relating to the medical condition of the generated virtual patient to a display device (not shown) connected to the data generation device 10.

[0065] Figure 5 is an example of a display screen showing time-series data related to a patient's medical condition. The example display screen in Figure 5 shows the changes in each patient's medical condition over time since they were diagnosed with stage 0 cancer. In the example display screen in Figure 5, the horizontal axis of the graph represents the time elapsed since the diagnosis of stage 0 cancer. The vertical axis of the graph represents the medical condition stage for each virtual patient. By referring to the example display screen in Figure 5, it is possible to understand, for example, the trends in the changes in each patient's medical condition.

[0066] Figure 6 shows an example of a display screen that shows the predicted results of a patient's condition. In the example display screen in Figure 6, the patient's name and the name of the disease they are suffering from are displayed at the top. In addition, the example display screen in Figure 6 shows a graph that shows the patient's current condition and the predicted results of the progression of the condition. In the graph showing the predicted results of the progression of the condition, the horizontal axis represents the elapsed time from the present, and the vertical axis represents the stage of the condition. In addition, the example display screen in Figure 6 shows the expected condition for each stage. By referring to the example display screen in Figure 6, users of the data generation device 10 can, for example, create a treatment plan or clinical trial plan based on the patient's condition if the condition progresses.

[0067] Figure 7 shows an example of a display screen that shows the predicted number of patients for each stage of the disease. In the example display screen of Figure 7, the name of the hospital being predicted is shown at the top. The example display screen of Figure 7 shows the predicted number of patients for each stage of the disease as time progresses. The example display screen of Figure 7 shows the predicted results of how the current number of patients for each stage of the disease, which is shown as the current value, will change over time. By referring to the example display screen of Figure 7, users of the data generation device 10 can, for example, create a patient admission plan or a clinical trial plan.

[0068] The storage unit 17 stores, for example, data relating to the process of generating time-series data relating to the medical condition of a virtual patient. The storage unit 17 stores, for example, data relating to the treatment of a patient. The storage unit 17 stores, for example, data relating to the treatment of a virtual patient. The storage unit 17 stores, for example, time-series data relating to the medical condition of a virtual patient. The storage unit 17 stores, for example, a generation model. The storage unit 17 stores, for example, a similarity calculation model. The storage unit 17 stores, for example, a selection model. The storage unit 17 stores, for example, a calculation model. The storage unit 17 stores, for example, a prediction model. The generation model, similarity calculation model, selection model, calculation model, and prediction model may be stored in storage means outside the data generation device 10.

[0069] The terminal device 20 is, for example, a terminal device used by the user in the process of extracting time-series data related to the medical condition of a virtual patient. The terminal device 20 outputs time-series data related to the medical condition of a virtual patient from the output unit 16 of the data generation device 10. The terminal device 20 then outputs time-series data related to the medical condition of a virtual patient to, for example, a display device (not shown).

[0070] The terminal device 20 may obtain prediction results for data regarding the treatment of target patients at a later stage than the initial stage from the output unit 16 of the data generation device 10. If prediction results for data regarding the treatment of target patients at a later stage than the initial stage are obtained, the terminal device 20 outputs the prediction results for data regarding the treatment of target patients at a later stage than the initial stage to a display device (not shown).

[0071] The terminal device 20 acquires information specifying the target disease, for example, which is input by the user. The terminal device 20 then outputs the information specifying the target disease to the acquisition unit 11 of the data generation device 10, for example.

[0072] Users are, for example, individuals engaged in healthcare-related work. For example, users are healthcare professionals. Healthcare professionals include, for example, doctors, nurses, pharmacists, laboratory technicians, clinical therapists, psychotherapists, caregivers, counselors, or administrative staff of medical institutions. Healthcare professionals are not limited to those listed above.

[0073] When time-series data on the medical condition of a virtual patient is used in a clinical trial, the users are, for example, staff members of a medical institution or staff members of an institution contracted by a medical institution to handle the clinical trial. An institution to which a hospital has contracted to handle the clinical trial is, for example, an SMO (Site Management Organization). A staff member of an institution contracted by a medical institution to handle the clinical trial is, for example, a CRC (Clinical Research Coordinator). The terminal device 20 may also be a terminal device used by staff members who conduct clinical trials at a pharmaceutical company or staff members of an institution contracted by a pharmaceutical company to conduct clinical trials of a drug. An institution contracted by a pharmaceutical company to conduct clinical trials is, for example, a CRO (Contract Research Organization). A staff member of an institution contracted by a pharmaceutical company to conduct clinical trials is, for example, a CRA (Clinical Research Associate). This is a terminal device used by staff members of a medical institution or staff members of an institution contracted by a medical institution to handle the clinical trial. An institution to which a hospital has contracted to handle the clinical trial is, for example, an SMO (Site Management Organization). A person in charge at an institution that has been contracted by a medical institution to perform tasks related to a clinical trial is, for example, a CRC (Clinical Research Coordinator). Alternatively, terminal device 20 may be a terminal device used by a person in charge of conducting clinical trials at a pharmaceutical company, or by a person in charge at an institution that has been contracted by a pharmaceutical company to perform clinical trials on a drug. An institution that has been contracted by a pharmaceutical company to perform clinical trials is, for example, a CRO (Contract Research Organization). Furthermore, a person in charge at an institution that has been contracted by a pharmaceutical company to perform clinical trials is, for example, a CRA (Clinical Research Associate).

[0074] The terminal device 20 can be, for example, a personal computer, a tablet computer, a smartphone, or a smartwatch. The information processing device used in the terminal device 20 is not limited to those mentioned above.

[0075] The data management device 30 is a device that stores data related to the patient's treatment. This data may include, for example, data from an electronic medical record entered by a physician. The electronic medical record information may also include data entered by a nurse, laboratory technician, physical therapist, or counselor. The patient's treatment data may also include information on one or more items other than those listed in the electronic medical record, such as the patient's disease information, comorbidity information, biomarkers, disease status, guideline score, medical history, effectiveness, and test results. The data management device 30 outputs the treatment data to, for example, the acquisition unit 11 of the data generation device 10.

[0076] The data management device 30 may store, for example, data related to patient treatment as anonymized information or pseudonymized information. Anonymized information is, for example, information that has been processed so that an individual cannot be identified even when cross-referenced with other information. Pseudonymized information is, for example, information that cannot identify an individual on its own, but can identify an individual when cross-referenced with other information.

[0077] The operation of the data generation device 10 in the process of generating time-series data on the medical condition of a virtual patient will be described. Figure 8 shows an example of the flow in the process in which the data generation device 10 generates time-series data on the medical condition of a virtual patient.

[0078] The acquisition unit 11 acquires data related to the patient's treatment (step S11). The acquisition unit 11 acquires data related to the patient's treatment from, for example, the data management device 30.

[0079] Once data regarding the patient's treatment is acquired, the virtual patient data generation unit 12 generates treatment data for each stage of the virtual patient's condition based on the patient's treatment data (step S12). For example, the virtual patient data generation unit 12 generates treatment data for each virtual patient.

[0080] When data related to the treatment of the virtual patient is generated, the time-series data generation unit 14 generates time-series data related to the condition of the virtual patient from the generated data related to the treatment of the virtual patient (step S13).

[0081] When time-series data regarding the virtual patient's medical condition is generated, the output unit 16 outputs the generated time-series data regarding the virtual patient's medical condition (step S14). The output unit 16 outputs the time-series data regarding the virtual patient's medical condition to, for example, the terminal device 20.

[0082] Each process in the data generation device 10 may be distributed and executed across multiple information processing devices connected via a network. For example, the processes in the virtual patient data generation unit 12 and the selection unit 13, and the processes in the time-series data generation unit 14 may be performed in separate information processing devices. The specific information processing devices on which each process in the data generation device 10 is performed can be configured as appropriate.

[0083] The data generation device 10 generates treatment data for each stage of the virtual patient's condition based on data related to the patient's treatment. Then, the data generation device 10 generates time-series data related to the virtual patient's condition from the generated treatment data. In this way, the data generation device 10 can easily generate time-series data related to the patient's condition by generating treatment data for the virtual patient and then generating time-series data related to the virtual patient's condition from the generated data.

[0084] Even in the case of rare diseases, which have few reported cases, the data generation device 10 can easily generate time-series data on the disease state of a hypothetical patient, for example, by generating time-series data on the disease state based on data on the treatment of the hypothetical patient. Therefore, by referring to time-series data on the disease state of a hypothetical disease, healthcare professionals, for example, can easily make decisions regarding the treatment of patients suffering from rare diseases. Furthermore, by referring to time-series data on the disease state of a hypothetical disease, those responsible for conducting clinical trials of drugs for rare diseases, for example, can improve the accuracy of their clinical trial plans.

[0085] Furthermore, by generating time-series data on the disease state of a hypothetical patient whose treatment data in the initial stages is similar to that of the target patient, the data generation device 10 can, for example, easily grasp the progression of the target patient's disease state.

[0086] Each process in the data generation 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 to perform each process in the data generation device 10. The computer 100 includes a CPU (Central Processing Unit) 101, memory 102, storage device 103, input / output interface (I / F) 104, and communication interface (I / F) 105.

[0087] The CPU 101 reads and executes computer programs that perform various processes from the storage device 103. The CPU 101 may be composed of a combination of multiple CPUs. Alternatively, the CPU 101 may be composed of a combination of a CPU and another type of processor. For example, the CPU 101 may be composed of a combination of a CPU and a GPU (Graphics Processing Unit). The memory 102 is composed of DRAM (Dynamic Random Access Memory) or the like, and temporarily stores computer programs executed by the CPU 101 and data being processed. The storage device 103 stores computer programs executed by the CPU 101. The storage device 103 is composed of, for example, a non-volatile semiconductor storage device. Other storage devices such as hard disk drives may be used for the storage device 103. The input / output interface 104 is an interface that receives input from the operator and outputs display data, etc. The communication interface 105 is an interface that sends 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 can also be configured similarly to the computer 100.

[0088] The computer programs 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 include magnetic tapes for data recording and magnetic disks such as hard disks. Optical discs such as CD-ROMs (Compact Disc Read Only Memory) can also be used as recording media. Non-volatile semiconductor memory devices may also be used as recording media.

[0089] Some or all of the above embodiments may also be described as follows, but are not limited to the following:

[0090] [Note 1] A means of acquiring data related to patient treatment, A virtual patient data generation means generates data on treatment at each stage of the virtual patient's condition based on data on the treatment of the aforementioned patient, A time-series data generation means generates time-series data relating to the medical condition of the virtual patient from the generated data relating to the treatment of the virtual patient, Output means for outputting time-series data relating to the medical condition of the generated virtual patient A data generation device equipped with the following features.

[0091] [Note 2] The virtual patient data generation means generates treatment data for each virtual patient using a generation model that generates treatment data for virtual patients from treatment data. The data generation device described in Appendix 1.

[0092] [Note 3] The virtual patient data generation means generates data from the virtual patient's electronic medical record as data related to the virtual patient's treatment, based on the data from the patient's electronic medical record, which is data related to the patient's treatment, using the generation model. The data generation device described in Appendix 2.

[0093] [Note 4] The virtual patient data generation means generates data relating to the virtual patient's complications as data relating to the virtual patient's treatment. A data generation device as described in any of the appendices 1 to 3.

[0094] [Note 5] The system further includes a selection means for selecting data from the generated data relating to the treatment of the virtual patient that is suitable as data for the target disease. A data generation device as described in any of the appendices 1 to 4.

[0095] [Note 6] The aforementioned acquisition means further acquires data regarding the patient's health condition before visiting the hospital as data regarding the patient's treatment. The virtual patient data generation means generates data relating to the treatment of the virtual patient at each stage of the illness, including before visiting a hospital. A data generation device as described in any of the appendices 1 to 5.

[0096] [Note 7] The aforementioned acquisition means further acquires data regarding the patient's health status after discharge as data regarding the patient's treatment. The virtual patient data generation means generates data related to treatment at each stage of the virtual patient's condition, including after discharge. A data generation device as described in any of the appendices 1 to 6.

[0097] [Note 8] The system further comprises a prediction means for predicting data on the treatment of the target patient at a later stage, using a machine learning model that predicts data on treatment at a later stage than the initial stage based on data on the patient's initial stage in a time series. A data generation device as described in any of the appendices 1 to 7.

[0098] [Note 9] The data relating to the aforementioned treatment includes the patient's perception of their medical condition. A data generation device as described in any of the appendices 1 to 8.

[0099] [Note 10] The time-series data regarding the condition of the aforementioned hypothetical patient represents the patient journey. A data generation device as described in any of the appendices 1 through 9.

[0100] [Note 11] The aforementioned generative model is a large-scale language model. A data generation device as described in Appendix 2 or 3.

[0101] [Note 12] The acquisition means acquires data regarding the patient's health condition before visiting the hospital, as well as data regarding the patient's treatment when the patient visited another hospital. The data generation device described in Appendix 6.

[0102] [Note 13] We obtain data on patient treatment, Based on the data regarding the treatment of the aforementioned patient, data regarding the treatment at each stage of the virtual patient's condition is generated. From the generated data regarding the treatment of the virtual patient, time-series data regarding the patient's condition is generated. Output time-series data regarding the medical condition of the generated virtual patient. Data generation method.

[0103] [Note 14] The process of acquiring data related to patient treatment, A process to generate data on treatment for each stage of the virtual patient's condition based on the data on the patient's treatment, A process to generate time-series data relating to the medical condition of the virtual patient from the generated data relating to the treatment of the virtual patient, A process to output time-series data regarding the medical condition of the generated virtual patient. A data generation program that causes a computer to execute a command.

[0104] Furthermore, some or all of the configurations described in Appendices 2 to 12, which are dependent on Appendice 1 above, may also be dependent on Appendices 13 and 14 in the same way as those described in Appendices 2 to 12. Moreover, not limited to Appendices 1, 13, and 14, some or all of the configurations described as appendices may also be dependent on various hardware, software, various recording means for recording software, or systems, without departing from the embodiments described above.

[0105] Although the present disclosure has been described above with reference to embodiments, the present disclosure is not limited to the forms of implementation described above. Various modifications to the structure and details of the present disclosure are possible, as can be understood by those skilled in the art within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate. [Explanation of symbols]

[0106] 10 Data generation device 11 Acquisition Department 12. Virtual Patient Data Generation Unit 13 Selection Department 14. Time-series data generation unit 15 Prediction Section 16 Output section 17 Memory section 20 Terminal devices 30 Data Management Devices 100 Computers 101 CPU 102 memory 103 Storage device 104 Input / Output Interfaces 105 Communication I / F

Claims

1. A means of acquiring data related to patient treatment, A virtual patient data generation means generates data on treatment at each stage of the virtual patient's condition based on data on the treatment of the aforementioned patient, A time-series data generation means generates time-series data relating to the medical condition of the virtual patient from the generated data relating to the treatment of the virtual patient, Output means for outputting time-series data relating to the medical condition of the generated virtual patient A data generation device equipped with the following features.

2. The virtual patient data generation means generates data related to the treatment of each virtual patient using a generation model that generates data related to the treatment of virtual patients from data related to the treatment of patients. The data generation apparatus according to claim 1.

3. The virtual patient data generation means generates data from the virtual patient's electronic medical record as data related to the virtual patient's treatment, using the generation model, based on the data from the patient's electronic medical record, which is data related to the patient's treatment. The data generation apparatus according to claim 2.

4. The virtual patient data generation means generates data relating to the virtual patient's complications as data relating to the virtual patient's treatment. A data generation device according to any one of claims 1 to 3.

5. The system further includes a selection means for selecting data from the generated data relating to the treatment of the virtual patient that is suitable as data for the target disease. A data generation device according to any one of claims 1 to 3.

6. The aforementioned acquisition means further acquires data regarding the patient's health condition before visiting the hospital as data regarding the patient's treatment. The virtual patient data generation means generates data relating to the treatment of the virtual patient at each stage of the illness, including before visiting a hospital. A data generation device according to any one of claims 1 to 3.

7. The aforementioned acquisition means further acquires data regarding the patient's health status after discharge as data regarding the patient's treatment. The virtual patient data generation means generates data related to treatment at each stage of the virtual patient's condition, including after discharge. A data generation device according to any one of claims 1 to 3.

8. The system further comprises a prediction means for predicting data on the treatment of the target patient at a later stage, using a machine learning model that predicts data on treatment at a later stage than the initial stage based on data on the patient's initial stage in a time series. A data generation device according to any one of claims 1 to 3.

9. We obtain data on patient treatment, Based on the data regarding the treatment of the aforementioned patient, data regarding the treatment at each stage of the virtual patient's condition is generated. From the generated data regarding the treatment of the virtual patient, time-series data regarding the patient's condition is generated. Output time-series data regarding the medical condition of the generated virtual patient. Data generation method.

10. The process of acquiring data related to patient treatment, A process to generate data on treatment for each stage of the virtual patient's condition based on the data on the patient's treatment, A process to generate time-series data regarding the medical condition of the virtual patient from the generated data regarding the treatment of the virtual patient, A process to output time-series data regarding the medical condition of the generated virtual patient. A data generation program that causes a computer to execute a command.

Images