Medical planning support system, medical planning support method, and medical planning support program

Abstract

A medical planning assistance system 40 provides assistance to a medical worker in performing an appropriate medical practice according to the condition of a patient by being provided with: an acquisition unit that acquires patient condition information 400 about the condition of a first patient; an inference model 43 that is a model obtained by learning of a relation among a condition 431 of a second patient, a medical practice 432 performed on the second patient, and a condition change 433 of the second patient due to the medical practice performed on the second patient; an inference unit 42 that infers recommended medical practice 420 to be recommended for the first patient, on the basis of the inference model 43 and the patient condition information 400 acquired by the acquisition unit 41; and an output unit 44 that outputs the recommended medical practice 420 inferred by the inference unit 42.

Description

[Technical field]

[0001] The present invention relates to a medical planning support system, a medical planning support device, a medical planning support method, and a recording medium storing a medical planning support program. [Background technology]

[0002] It is expected that the rapidly evolving technology of machine learning will be utilized to support people working in various fields, including the medical field.

[0003] Patent Document 1 discloses a medical data structure including a patient's medical data consisting of multiple elements and morbidity information for identifying whether the patient is a sick person who has been hospitalized or a normal person who has never been hospitalized. This data structure is used in a process of performing supervised learning to change the weight of any element of the medical data according to a weighting rule to change the tensor data and generate a core tensor similar to a target core tensor when the patient corresponds to a recovered patient.

[0004] Patent Document 2 discloses a learning program that generates a hypothesis set that lists hypotheses that satisfy a specific condition, the hypothesis set being configured by a combination of explanatory variables from training data, each of which has explanatory variables and a target variable. This learning program performs learning to calculate the weight of each of a plurality of hypotheses included in the hypothesis set, based on whether each of the hypotheses is established for each of the training data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2020-101901 A [Patent Document 2] JP 2020-046888 A [Non-patent literature]

[0006] [Non-Patent Document 1] Lu Wang, Wenchao Yu, Wei Wang, Wei Cheng, Martin Renqiang Min, Bo Zong, Xiaofeng He, Hongyuan Zha, Wei Zhang, Haifeng Chen, "Adversarial Cooperative Imitation Learning for Dynamic Treatment Regimes", WWW '20, April 20-24, 2020, Taipei, Taiwan, Pages 1785-1795 Summary of the Invention [Problem to be solved by the invention]

[0007] In the medical field, medical treatment for a patient in a certain condition is usually decided based on the knowledge and experience of each doctor. For this reason, the accuracy of the criteria for deciding on a medical treatment is insufficient, and for example, if a medical professional such as a doctor with insufficient knowledge and experience performs an inappropriate medical treatment, the patient's condition may not improve or may worsen. Patent documents 1 and 2 do not mention this problem.

[0008] A primary object of the present invention is to provide a medical planning support system and the like that can support medical personnel so as to perform appropriate medical treatment according to the condition of a patient. [Means for solving the problem]

[0009] A medical planning support system according to one embodiment of the present invention comprises an acquisition means for acquiring patient condition information relating to the condition of a first patient, an estimation means for estimating a recommended medical procedure to be recommended for the first patient based on the patient condition information acquired by the acquisition means and an estimation model, and an output means for outputting the recommended medical procedure estimated by the estimation means, wherein the estimation model is a model that has learned the relationship between the condition of a second patient, medical procedures performed on the second patient, and changes in the condition of the second patient due to the medical procedures performed on the second patient.

[0010] In another aspect of achieving the above-mentioned object, a medical planning support method according to one embodiment of the present invention includes an information processing device that acquires patient condition information regarding the condition of a first patient, estimates a recommended medical procedure for the first patient based on the acquired patient condition information and an estimation model, and outputs the estimated recommended medical procedure, wherein the estimation model is a model that has learned the relationship between the condition of a second patient, the medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient.

[0011] In addition, in a further aspect of achieving the above-mentioned object, a medical planning support program according to one embodiment of the present invention is a program for causing a computer to execute an acquisition process of acquiring patient condition information regarding the condition of a first patient, an estimation process of estimating a recommended medical action to be recommended for the first patient based on the patient condition information acquired by the acquisition process and an estimation model, and an output process of outputting the recommended medical action estimated by the estimation process, wherein the estimation model is a model that has learned the relationship between the condition of a second patient, medical actions performed on the second patient, and changes in the condition of the second patient due to the medical actions performed on the second patient.

[0012] Furthermore, the present invention can also be realized by a computer-readable, non-volatile recording medium in which such a medical care planning support program (computer program) is stored. Effect of the Invention

[0013] According to the present invention, it is possible to obtain a medical plan support system and the like that can support medical staff so that appropriate medical treatment is performed according to the condition of a patient. [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a block diagram showing a configuration of a medical plan support system 10 according to a first embodiment of the present invention. [Diagram 2] 1 is a diagram illustrating patient condition information 101 according to a first embodiment of the present invention. [Diagram 3] 1 is a diagram illustrating medical treatment information 102 according to the first embodiment of the present invention. [Figure 4] 2 is a diagram illustrating an example of status change information 103 according to the first embodiment of the present invention. FIG. [Diagram 5] 1 is a flowchart showing an operation (processing) of the medical planning support system 10 according to the first embodiment of the present invention for generating an estimation model 140 (performing machine learning). [Figure 6] FIG. 2 is a diagram illustrating an example of an aspect in which the output unit 16 according to the first embodiment of the present invention displays patient condition information 101 and medical treatment information 102 regarding a patient undergoing treatment, and an estimated recommended medical treatment 104 on a display screen 200. [Figure 7] 4 is a flowchart showing an operation of the medical plan support system 10 according to the first embodiment of the present invention for estimating a recommended medical action 104. [Figure 8] FIG. 11 is a block diagram showing a configuration of a medical plan support system 40 according to a second embodiment of the present invention. [Figure 9] FIG. 1 is a block diagram showing a configuration of an information processing system 900 capable of realizing the medical plan support system 10 according to the first embodiment or the medical plan support system 30 according to the second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] A system, which is an example of an embodiment described later, uses a trained model (also called an estimation model) generated by machine learning (e.g., deep learning) when estimating a target event from certain input information. In addition, the system may use existing machine learning techniques such as ACIL (Adversarial Cooperative Imitation Learning) and SAiL (Skill Acquisition Learning) shown in Non-Patent Document 1.

[0016] ACIL is a technology that applies AIL (Adversarial Imitation Learning), a technology that inputs past successful cases and imitates the behavior in the successful cases, and avoids failures by also inputting failure cases. SAiL is a technology that has multiple behavior imitators and selects a behavior imitator according to the situation, thereby accurately imitating (learning) the behavior of an expert. Note that detailed explanations of these technologies will be omitted in the embodiments described below.

[0017] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

[0018] <First embodiment> 1 is a block diagram showing the configuration of a medical plan support system 10 according to a first embodiment of the present invention. The medical plan support system 10 is a system that supports medical professionals so that they can decide on appropriate medical procedures according to the condition of a patient. The medical professionals include doctors who perform medical procedures, as well as nurses, physical therapists, occupational therapists, radiologists, and the like who support doctors according to their instructions.

[0019] The medical planning support system 10 generates a trained model (also referred to as an estimation model) using information on the condition of a patient and the medical procedures performed on the patient for various past treatment cases labeled as success cases in which the patient's condition improved or failure cases in which the patient's condition did not improve. The medical planning support system 10 uses the trained model to estimate medical procedures recommended to be performed on a patient according to the patient's condition during treatment. The medical planning support system 10 is composed of at least one or more information processing devices.

[0020] A management terminal device 20 (also referred to as a display device) and medical equipment 30 are communicatively connected to the medical planning support system 10. The management terminal device 20 is, for example, a personal computer or other information processing device that is used when a medical professional using the medical planning support system 10 inputs information to the medical planning support system 10 and checks information output from the medical planning support system 10. The management terminal device 20 is provided with a display screen 200 that displays the information output from the medical planning support system 10.

[0021] The medical device 30 is, for example, a device that measures a patient's condition such as body temperature or blood pressure, or a device that can adjust the amount of medicine administered to the patient. The medical planning support system 10 can collect data indicating the patient's condition from the medical device 30. Alternatively, the medical planning support system 10 can control the operation of the medical device 30 when a doctor performs a medical procedure using the medical device 30.

[0022] The medical planning support system 10 includes an acquisition unit 11, a classification unit 12, an analysis unit 13, a model generation unit 14, an estimation unit 15, and an output unit 16. The acquisition unit 11, the classification unit 12, the analysis unit 13, the model generation unit 14, the estimation unit 15, and the output unit 16 are examples of an acquisition means, a classification means, an analysis means, a model generation means, an estimation means, and an output means, respectively.

[0023] Next, the operation of the medical plan support system 10 according to this embodiment to generate or update (re-learn) an estimation model 140 for estimating a recommended medical action 104, which is a medical action recommended to be performed on a patient undergoing treatment (hereinafter also referred to as a "target patient"), will be described. After that, the operation of estimating the recommended medical action 104 for the patient using the estimation model 140 will be described.

[0024] <Operation of generating (and updating or re-learning) the estimation model 140> First, we will explain the operation of the medical plan support system 10 according to this embodiment from acquiring information for treatment to generating or updating the estimation model 140 for estimating medical procedures recommended for a patient undergoing treatment.

[0025] The acquisition unit 11 acquires the patient condition information 101, the medical treatment information 102, and the status change information 103 related to past treatment cases to be learned from a computer device or a database (not shown) via a network. The acquisition unit 11 may acquire the patient condition information 101, the medical treatment information 102, and the status change information 103, for example, periodically. The acquisition unit 11 may acquire the patient condition information 101, the medical treatment information 102, and the status change information 103, for example, in response to an instruction input by a medical professional via the management terminal device 20.

[0026] The acquisition unit 11 includes, for example, a communication circuit connected to one or more computer devices or databases that transmit the patient condition information 101, the medical treatment information 102, and the condition change information 103, and a storage device that stores the information acquired by the communication circuit. The storage device may be a hard disk 904 or a RAM 903 of an information processing system 900 shown in Fig. 9, which will be described later.

[0027] The patient condition information 101 is information that represents the history of the condition of each patient. The patient condition is, for example, the blood condition such as body temperature, heart rate, blood pressure, blood oxygen concentration, and the amount and concentration of substances contained in the blood, the urine condition such as the color, odor, and contained substances of urine, the stool condition, and the like. The patient condition is, for example, the consciousness state such as the consciousness level, the pain state such as the presence or absence of pain and the degree of pain, the movement state of the joint such as the range of motion of the joint, and the like obtained when the doctor diagnoses the patient. The patient condition is, for example, the organ condition evaluated by the doctor from an image obtained by a medical device such as an X-ray. The patient condition information is not limited to the above, as long as it is information regarding the health condition or biological condition of the patient.

[0028] The above-mentioned patient condition information may be obtained during a specific examination such as a health check, may be obtained during medical treatment or diagnosis, or may be obtained by a wearable device.

[0029] Fig. 2 is a diagram illustrating an example of data of patient condition information 101 according to this embodiment. The patient condition information 101 represents the history (progression) of the condition of each patient, such as patient A and patient B. The patient condition information 101 illustrated in Fig. 2 represents the patient's condition such as body temperature, heart rate, blood pressure, and blood glucose level, but the patient's condition represented by the patient condition information 101 is not limited to these.

[0030] In addition, while the patient condition information 101 illustrated in FIG. 2 represents the patient's condition using specific numerical values, the patient condition information 101 may represent the patient's condition, for example, by indicating whether each item is a normal value or an abnormal value.

[0031] The medical treatment information 102 is information that represents the history of medical treatments performed on each patient. For example, medical treatments include surgery, drug therapy, radiation therapy, rehabilitation therapy, etc. In the case of surgery, for example, the medical treatment information 102 represents the part of the body that was operated on and the method of surgery, etc. The method of surgery is, for example, laparotomy, catheter surgery, etc. In the case of drug therapy, for example, the medical treatment information 102 represents the type (name, etc.) of the administered medicine, the method of use, the dosage, etc. In the case of radiation therapy, for example, the medical treatment information 102 represents the type of radiation used, such as X-rays, gamma rays, electron beams, etc., the irradiated area, the irradiation time, etc. In the case of rehabilitation therapy, for example, the medical treatment information 102 represents the part of the body that was treated and the specific type of rehabilitation, etc.

[0032] Fig. 3 is a diagram illustrating medical treatment information 102 according to this embodiment. The medical treatment information 102 represents a history of medical treatments performed for each patient, such as patient A and patient B. The medical treatment information 102 illustrated in Fig. 3 represents that patient A is receiving continuous drug treatment.

[0033] The status change information 103 is information that represents a history of changes in the status of each patient. A change in status represents, for example, whether the status of the patient has improved or worsened, that is, whether the treatment has been successful or unsuccessful. Whether the status of the patient has improved or worsened can be determined based on a determination criterion defined using a predetermined KPI (Key Performance Indicator). In addition to whether the status has improved or worsened, if the KPI has not improved or worsened by a predetermined value or more, it may be indicated that the status of the patient has not changed, such as "maintained as is."

[0034] As the KPI, various indices that indicate the above-mentioned patient's condition (for example, body temperature, blood pressure, blood sugar level, etc.) can be used. The judgment criteria may be, for example, that when these indices rise above a predetermined reference value, it indicates that the condition has worsened, and when these indices fall below the predetermined reference value, it indicates that the condition has improved.

[0035] FIG. 4 is a diagram illustrating the status change information 103 according to the present embodiment. The status change information 103 represents a history of improvement or deterioration of the status of each patient, such as patient A and patient B. According to the status change information 103 illustrated in FIG. 4, patient A developed a fever at 11:00 on August 24, 2020, and his condition worsened. For this reason, as shown in the medical procedure information 102 illustrated in FIG. 3, a medical procedure was performed on patient A at 11:10 on August 24, 2020 to administer two units of drug Z. As a result, as shown in the patient status information 101 illustrated in FIG. 2 and the status change information 103 illustrated in FIG. 4, patient A's body temperature dropped to normal at 19:00 on August 24, 2020, and his condition improved. The above-mentioned examples illustrated in FIGS. 2 to 4 show that the treatment of administering two units of drug Z to patient A was successful.

[0036] The acquiring unit 11 may generate the condition change information 103 from the patient condition information 101, without acquiring it from an external source. The acquiring unit 11 can determine the change in the patient's condition from the history of the patient's condition indicated by the patient condition information 101. The acquiring unit 11 may then determine whether the patient's condition has improved or worsened based on a determination criterion defined using the above-mentioned predetermined KPI, which is given in advance.

[0037] The acquisition unit 11 stores the patient condition information 101, medical treatment information 102, and condition change information 103 acquired as described above in a storage device (not shown) (for example, a memory or a hard disk).

[0038] The model generation unit 14 uses the patient condition information 101, medical treatment information 102, and condition change information 103 acquired by the acquisition unit 11 as training data to generate an estimation model 140 (trained model) used by the estimation unit 15 described later when estimating a recommended medical treatment 104 through machine learning. The estimation model 140 is generated by learning the relationship between the past condition of the patient, the medical treatment performed on the patient in the past, and the change in the past condition of the patient due to the performed medical treatment. For example, the model generation unit 14 generates or updates the estimation model 140 through the above-mentioned machine learning techniques ACIL and SAiL. The model generation unit 14 performs machine learning to generate or update the estimation model 140 using the above-mentioned training data through a processor.

[0039] The treatment cases to be learned, represented by patient condition information 101, medical treatment information 102, and condition change information 103, are labeled as success cases or failure cases according to the determination result of whether the patient's condition worsened or improved, as indicated by the condition change information 103. The treatment case in which two units of drug Z were administered to patient A with a fever described above is a success case, as shown in Figures 2 to 4.

[0040] In addition, for example, in cases where a patient's condition changes (improves or worsens) a long time after a certain medical procedure, it is difficult to determine the causal relationship between the medical procedure and the patient's condition. For such cases, for example, proven research data that has been judged by experts to be successful or unsuccessful from a long-term perspective can be used as training data.

[0041] When generating or updating the estimation model 140, the model generation unit 14 can use the results of estimation of medical procedures recommended for past patients by the estimation unit 15, which will be described later, regarding the teacher data.

[0042] 2 and 3, the estimation unit 15 estimates a medical procedure recommended for a patient who is receiving drug therapy with drugs X and Y and develops a fever. In this case, as described above, there is a record that the patient's condition improved by administering two units of drug Z, but the estimation unit 15 estimates the recommended medical procedure without referring to this record.

[0043] The model generation unit 14 judges whether the teacher data indicates a success case or a failure case based on the state change information 103. When the teacher data indicates a success case, the model generation unit 14 generates or updates the estimation model 140 so that the estimation result of the recommended medical procedure by the estimation unit 15 becomes closer (similar) to the actually performed medical procedure indicated by the medical procedure information 102.

[0044] For example, assume that the estimation unit 15 estimates that the recommended medical practice for the teacher data indicating a successful case is to administer one unit of drug Z. In this case, the model generation unit 14 generates or updates the estimation model 140 so that the estimation unit 15 can estimate that the recommended medical practice is to administer two units of drug Z, which was actually performed. That is, the model generation unit 14 generates or updates the estimation model 140 so that the estimation unit 15 can estimate a medical practice that is closer (similar) to the medical practice performed in the successful case.

[0045] In addition, when the training data indicates a failure case opposite to the above-mentioned example, the model generation unit 14 generates or updates the estimation model 140 so that the estimation result of the recommended medical procedure by the estimation unit 15 becomes more distant (different) from the actually performed medical procedure indicated by the medical procedure information 102.

[0046] For example, suppose that a patient undergoing drug treatment with drugs X and Y shown in Figures 2 to 4 develops a fever, and the patient is administered two units of drug Z, but the patient's condition does not improve. In other words, the training data in this case shows a failure case.

[0047] For the teacher data showing such a failure case, the estimation unit 15 estimates, for example, administration of one unit of drug Z as the recommended medical action. In this case, the model generation unit 14 generates or updates the estimation model 140 so as not to estimate, as the recommended medical action, a medical action similar to the administration of two units of drug Z that was actually performed. That is, the model generation unit 14 generates or updates the estimation model 140 so that the estimation unit 15 estimates, for example, a medical action of administering a drug other than drug Z as the recommended medical action.

[0048] The model generation unit 14 stores the estimation model 140 generated or updated as described above in a non-volatile storage device (not shown). The model generation unit 14 can gradually improve the estimation accuracy by updating (also called re-learning) the estimation model 140, for example, at predetermined time intervals.

[0049] 1 classifies the medical procedures indicated by the medical procedure information 102 acquired by the acquisition unit 11 into groups based on their similarities, and inputs the contents of the groups to the model generation unit 14. However, it is assumed that the classification criteria are provided in advance to the classification unit 12 by, for example, a medical professional. The classification unit 12 may input, for example, an identifier capable of identifying a group to the model generation unit 14.

[0050] For example, the classification unit 12 classifies into the same group a medical procedure using a first drug and a medical procedure using a second drug that has the same effect as the first drug but is of a different type such as a name, etc. More specifically, the first drug is a brand-name drug, and the second drug is a generic drug.

[0051] The classification unit 12 may classify medical procedures such as surgery, radiation therapy, rehabilitation therapy, and the like, in addition to such drug therapy, into groups based on the similarities between them.

[0052] The classification unit 12 may also classify medical procedures into groups having a hierarchical structure based on their similarity. For example, the classification unit 12 classifies medical procedures performed on cancer patients into large groups such as surgery, drug therapy, and radiation therapy. The classification unit 12 then classifies the medical procedures into smaller groups that belong to the large groups, for example, based on the type of surgery, the type of medicine used, the type of radiation used, etc.

[0053] When the model generation unit 14 receives the above-mentioned group from the classification unit 12, the model generation unit 14 generates or updates an estimation model 140 using the group as a medical procedure. In this case, the estimation unit 15 estimates the group of medical procedures as a recommended medical procedure based on the estimation model 140 generated using the group of medical procedures.

[0054] The analysis unit 13 shown in Fig. 1 analyzes whether or not the patient condition information 101, medical treatment information 102, and condition change information 103 acquired by the acquisition unit 11 indicate that a combination of multiple medical treatments has affected the patient's condition. For example, when multiple medical treatments are performed on a patient at the same timing and the patient's condition is found to have improved or worsened as a result, the analysis unit 13 determines that the combination of those medical treatments has affected the patient's condition. It is assumed that the analysis unit 13 is provided with a criterion for determining that the combination of multiple medical treatments has affected the patient's condition in advance, for example, by a medical professional.

[0055] The combination of multiple medical procedures is, for example, a drug therapy using a combination of multiple medicines. Alternatively, the combination of multiple medical procedures may be, for example, a combination of two or more medical procedures selected from surgery, drug therapy, radiation therapy, and rehabilitation therapy.

[0056] If the analysis unit 13 finds, as a result of the above analysis, that a certain combination of medical practices has an effect on the patient's condition, the analysis unit 13 inputs the combination to the model generation unit 14. A combination is data related to medical practices to be combined.

[0057] When the model generation unit 14 receives the above-mentioned combination from the analysis unit 13, the model generation unit 14 generates or updates the estimation model 140 as a medical procedure using the combination. In this case, the estimation unit 15 estimates the combination of medical procedures as a recommended medical procedure based on the estimation model 140 generated using the combination of medical procedures.

[0058] Next, the operation (processing) of the medical planning support system 10 according to this embodiment to generate the estimation model 140 (perform machine learning) will be described in detail with reference to the flowchart in FIG.

[0059] The acquisition unit 11 acquires patient condition information 101, medical treatment information 102, and condition change information 103 related to a patient previously treated (step S101). The estimation unit 15 estimates a medical treatment recommended for the patient based on the patient condition information 101, medical treatment information 102, and estimation model 140 related to the patient acquired by the acquisition unit 11 (step S102).

[0060] The model generating unit 14 checks whether the condition change information 103 indicates that the condition of the patient has improved (that is, whether the input treatment case is a success case or a failure case) (step S103).

[0061] If the condition change information 103 indicates that the patient's condition has improved (i.e., the input treatment case is a successful case) (Yes in step S104), the process proceeds to step S105. The model generation unit 14 generates or updates the estimation model 140 so that the difference between the estimated result of the medical procedure recommended for the patient and the performed medical procedure indicated by the medical procedure information 102 becomes smaller (step S105), and the entire process ends.

[0062] If the condition change information 103 indicates that the patient's condition has not improved (i.e., the input treatment case is a failed case) (Yes in step S104), the process proceeds to step S106. The model generation unit 14 generates or updates the estimation model 140 so that the difference between the estimated result of the medical treatment recommended for the patient and the performed medical treatment indicated by the medical treatment information 102 becomes larger (step S106), and the entire process ends.

[0063] <Operation to estimate recommended medical procedures for patients undergoing treatment> Next, the operation of the medical plan support system 10 according to this embodiment, which uses the generated or updated estimation model 140 to estimate a medical procedure that is recommended for a patient undergoing treatment, will be described.

[0064] The acquisition unit 11 acquires the patient condition information 101 and the medical treatment information 102 from an external device (not shown) in the same manner as when the medical plan support system 10 generates the estimation model 140. However, the acquisition unit 11 does not acquire this information as the above-mentioned teacher data, but acquires it as data on an estimation target related to a medical treatment recommended to be performed on a patient (also referred to as a first patient) under treatment.

[0065] As described above, it is assumed that the estimation model 140 has been generated based on the patient condition information 101, medical treatment information 102, and condition change information 103 related to a patient who was previously treated (also referred to as a second patient). In this case, the acquisition unit 11 acquires the patient condition information 101 and medical treatment information 102 related to the patient currently being treated, which are input by a medical professional such as a doctor via the management terminal device 20. The aspects of the patient condition information 101 and medical treatment information 102 related to the patient currently being treated are similar to the patient condition information 101 and medical treatment information 102 used to generate the estimation model 140 shown in Figs. 2 and 3.

[0066] FIG. 6 is a diagram illustrating an example of how the output unit 16 according to this embodiment displays, on a display screen 200, patient condition information 101 and medical procedure information 102 relating to a patient undergoing treatment, and a recommended medical procedure 104 estimated by the estimation unit 15.

[0067] The display screen 200 shown in Fig. 6 is displayed by the output unit 16 after a doctor among medical staff logs in to the medical plan support system 10 with his / her own ID (Identification) via the management terminal device 20. The output unit 16 displays the display screen 200 shown in Fig. 6 in response to an input operation by the doctor to select patient A on a screen (not shown) for selecting one of a plurality of patients.

[0068] The output unit 16 outputs at least one of patient condition information 101, medical treatment information 102, and condition change information 103 relating to the target patient.

[0069] For example, the output unit 16, as a display control unit, displays patient condition information 101 and medical treatment information 102 regarding patient A on a display screen 200, as exemplified in Fig. 6. The patient condition information 101 and the medical treatment information 102 may include information input by a doctor through an input operation on the display screen 200 exemplified in Fig. 6, information collected from a medical device 30, or information separately input by a medical professional such as a nurse via the management terminal device 20.

[0070] In the display screen 200 illustrated in FIG. 6, patient condition information 101 and medical treatment information 102 regarding patient A undergoing treatment, which have been input by a doctor, are stored by the medical planning support system 10 in a computer device or database that manages the patient condition information 101 and the medical treatment information 102.

[0071] The estimation unit 15 estimates a recommended medical action 104, which is a medical action recommended to be performed on a patient under treatment, based on the patient condition information 101 and medical action information 102 related to the patient under treatment acquired by the acquisition unit 11, and the estimation model 140. The estimation unit 15 also has a function of estimating a recommended medical action based on the patient condition information 101 and medical action information 102 input as teacher data, and the estimation model 140.

[0072] When the estimation model 140 is generated using a group of medical procedures by the above-described operation of the classification unit 12, the estimation unit 15 may estimate the group of medical procedures as the recommended medical procedures 104.

[0073] When the estimation model 140 is generated using a combination of medical procedures by the above-mentioned operation of the analysis unit 13, the estimation unit 15 may estimate the combination of medical procedures as the recommended medical procedure 104. For example, the recommended medical procedure 104 is a series of medical procedures including multiple medical procedures, such as "perform catheter surgery, and then administer drug X by infusion twice a day, in the morning and evening, for one week."

[0074] The estimation unit 15 outputs the estimated recommended medical action 104 to the output unit 16. At this time, the estimation unit 15 may also output the reason why the recommended medical action 104 was estimated to the output unit 16. The estimation unit 15 is equipped with an attention technique in deep learning, so that it can output the reason for the estimation.

[0075] The output unit 16 outputs the recommended medical action estimated by the estimation unit 15. For example, the output unit 16, as a display control unit, displays the recommended medical action 104 input from the estimation unit 15 on the display screen 200 of the management terminal device 20. That is, the output unit 16 controls the management terminal device 20 to display the recommended medical action 104 on the display screen 200 of the management terminal device 20. When the doctor completes an input operation of inputting the patient condition information 101 and medical action information 102 regarding patient A on the display screen 200, the output unit 16 may display the recommended medical action 104 on the display screen 200 as feedback for the input operation.

[0076] 6, the estimation unit 15 estimates a recommended medical action 104 indicating that "continue administering one unit each of drug X and drug Y at 9:00, and one unit of drug X at 20:00" for patient A who is receiving drug treatment with drug X and drug Y. Alternatively, for example, the estimation unit 15 estimates a recommended medical action 104 such as "perform rehabilitation treatment to stabilize the lower trunk and increase the range of motion of dorsiflexion of the ankle joint" for patient B who is receiving rehabilitation treatment.

[0077] In addition, as shown in FIG. 6, the output unit 16 displays an “Approve” button on the display screen 200 to prompt the doctor to input an input operation indicating whether or not the recommended medical action 104 displayed on the display screen 200 is approved.

[0078] For example, when the “Approve” button is selected by a doctor using a mouse or touch operation, the output unit 16 generates control information for the medical device 30 to be used by the doctor when performing the recommended medical procedure 104 displayed on the display screen 200, and transmits the generated control information to the medical device 30.

[0079] Next, the operation (processing) of the medical plan support system 10 according to this embodiment for estimating the recommended medical action 104 will be described in detail with reference to the flowchart of FIG.

[0080] Patient condition information 101 and medical treatment information 102 regarding a patient undergoing treatment are input via the management terminal device 20 (step S201). The estimation unit 15 estimates a recommended medical treatment 104 for the patient based on the input patient condition information 101 and medical treatment information 102 and the estimation model 140 (step S202).

[0081] The output unit 16 displays the recommended medical action 104 estimated by the estimation unit 15 on the display screen 200, and also displays information on the display screen 200 to prompt the doctor to input information indicating whether or not to approve the recommended medical action 104 (step S203).

[0082] If the doctor does not perform an input operation on the display screen 200 to approve the recommended medical practice 104 (No in step S204), the entire process ends. If the doctor performs an input operation on the display screen 200 to approve the recommended medical practice 104 (Yes in step S204), the output unit 16 generates control information for the medical device 30 that performs the recommended medical practice 104, outputs the generated control information to the medical device 30, and the entire process ends.

[0083] The medical plan support system 10 according to the present embodiment can support medical personnel to perform appropriate medical procedures according to the condition of the patient because the medical plan support system 10 estimates medical procedures recommended for a patient currently undergoing treatment based on the condition of the patient currently undergoing treatment and an estimation model 140 generated from past treatment cases.

[0084] The effects achieved by the medical plan support system 10 according to this embodiment will be described in detail below.

[0085] In the medical field, medical treatment for a patient with a certain condition is usually decided based on the knowledge and experience of each doctor. For this reason, the accuracy of the criteria for deciding on medical treatment is insufficient, and for example, if a medical professional such as a doctor with insufficient knowledge and experience performs an inappropriate medical treatment, the patient's condition may not improve or may worsen.

[0086] In response to such a problem, the medical plan support system 10 according to the present embodiment includes an acquisition unit 11, an estimation model 140, an estimation unit 15, and an output unit 16, and operates as described above with reference to, for example, FIGS. 1 to 7. That is, the acquisition unit 11 acquires patient condition information 101 relating to the condition of a patient (first patient) undergoing treatment. The estimation model 140 is a model that learns the relationship between the condition of a past patient (second patient), medical procedures performed on the past patient, and changes in the condition of the past patient due to the performed medical procedures. The estimation unit 15 estimates a recommended medical procedure 104 recommended for the patient undergoing treatment based on the patient condition information 101 relating to the condition of the patient undergoing treatment and the estimation model 140. This allows the medical plan support system 10 to appropriately support the treatment of a patient so that, for example, a medical professional such as a doctor who has insufficient knowledge and experience does not perform an inappropriate medical procedure.

[0087] The medical plan support system 10 according to this embodiment also includes a model generation unit 14 that generates or updates an estimation model 140 based on patient condition information 101, medical treatment information 102, and status change information 103 related to past patients. In the process of generating or updating the estimation model 140, the medical plan support system 10 estimates a medical treatment recommended for the past patient based on the patient condition information 101 and medical treatment information 102 related to the past patient and the estimation model 140 by the estimation unit 15. The model generation unit 14 updates the estimation model 140 so that the difference between the medical treatment actually performed in a successful treatment case and the medical treatment estimated by the estimation unit 15 becomes smaller. Alternatively, the model generation unit 14 updates the estimation model 140 so that the difference between the medical treatment actually performed in a failed treatment case and the medical treatment estimated by the estimation unit 15 becomes larger. In other words, the medical planning support system 10 updates the estimation model 140 by using both successful and unsuccessful treatment cases as training data, so that an estimation model 140 that realizes highly accurate estimations can be constructed in a short period of time.

[0088] In addition, when estimating a recommended medical action using the estimation model 140, there are the following problems. For example, there are multiple medical actions that have the same effect (i.e., are similar), just as there are drugs that have the same effect but different types (names, etc.), such as original drugs and generic drugs. If such multiple medical actions are treated as separate, the number of samples for each medical action will be reduced, which will cause a decrease in accuracy as teacher data. To address this issue, the medical plan support system 10 according to this embodiment includes a classification unit 12 that classifies medical actions into groups based on their similarity and inputs the groups to the model generation unit 14, and performs processing on a group basis in generating the estimation model 140 and estimating the recommended medical action 104. As a result, the medical plan support system 10 can solve the problem of the above-mentioned decrease in accuracy as teacher data.

[0089] In addition, when estimating a recommended medical action using the estimation model 140, there is another problem as follows. For example, in rehabilitation treatment, the effect of treatment may be obtained by combining and performing multiple medical actions (treatments). Alternatively, a combination of multiple medicines may cause side effects. Therefore, when focusing only on a single medical action, the influence of the combination of the above-mentioned multiple medical actions cannot be captured, which causes a decrease in accuracy as teacher data. To address this problem, the medical plan support system 10 according to this embodiment includes an analysis unit 13 that analyzes whether or not a combination of multiple medical actions has affected the patient's condition and inputs the combination that has affected the patient's condition to the model generation unit 14, and performs processing on a combination basis in generating the estimation model 140 and estimating the recommended medical action 104. As a result, the medical plan support system 10 can solve the problem of the above-mentioned decrease in accuracy as teacher data.

[0090] Furthermore, the medical plan support system 10 according to this embodiment displays a recommended medical action 104 as feedback when an input operation is performed on the display screen 200 to input the condition of a patient under treatment or the medical action performed. This enables the medical plan support system 10 to improve the work efficiency of medical professionals such as doctors.

[0091] Furthermore, the medical plan support system 10 according to this embodiment displays an "Approve" button on the display screen 200 to prompt the doctor to perform an input operation indicating whether or not to approve the recommended medical action 104. If the recommended medical action 104 is approved by the doctor, the medical plan support system 10 generates control information for the medical device 30 that performs the recommended medical action 104, and outputs the control information to the medical device 30. This allows the medical plan support system 10 to avoid the erroneous performance of an inappropriate medical action that has not been approved by the doctor.

[0092] <Second embodiment> 8 is a block diagram showing a configuration of a medical plan support system 40 according to a second embodiment of the present invention. The medical plan support system 40 includes an acquisition unit 41, an estimation unit 42 that uses an estimation model 43, and an output unit 44. Note that the acquisition unit 41, the estimation unit 42, and the output unit 44 are examples of an acquisition means, an estimation means, and an output means, respectively.

[0093] The acquisition unit 41 acquires patient condition information 400 relating to the condition of a first patient. The acquisition unit 41 operates in the same manner as the acquisition unit 11 according to the first embodiment, for example. The first patient is, for example, a patient currently undergoing treatment (i.e., a target patient) as in the first embodiment.

[0094] The estimation model 43 is a model that learns and generates a relationship between a state 431 of the second patient, a medical procedure 432 performed on the second patient, and a change 433 in the state of the second patient due to the medical procedure 432 performed on the second patient. The second patient is, for example, a past patient, as in the first embodiment. The estimation model 43 is, for example, a trained model that represents a result of machine learning performed on the relationship between the state 431 of the second patient, a medical procedure 432 performed on the second patient, and a change 433 in the state of the second patient due to the medical procedure 432 performed on the second patient, as in the estimation model 140 according to the first embodiment.

[0095] The second patient's condition 431 may be, for example, information similar to the patient condition information 101 described in the first embodiment with reference to Fig. 2. The medical procedure performed on the second patient 432 may be, for example, information similar to the medical procedure information 102 described in the first embodiment with reference to Fig. 3. The change in the second patient's condition 433 may be, for example, information similar to the condition change information 103 described in the first embodiment with reference to Fig. 4.

[0096] The estimation unit 42 estimates a recommended medical action 420 to be recommended for the first patient based on the patient condition information 400 acquired by the acquisition unit 41 and the estimation model 43. More specifically, the estimation unit 42 estimates the recommended medical action 420 by operating in a similar manner to the estimation unit 15 according to the first embodiment, for example.

[0097] The output unit 44 outputs the recommended medical action 420 estimated by the estimation unit 42. The output unit 44 operates in the same manner as the output unit 16 according to the first embodiment, for example.

[0098] The medical plan support system 40 according to this embodiment can support medical personnel to perform appropriate medical procedures according to the condition of the patient. This is because the medical plan support system 40 estimates medical procedures recommended for a patient currently undergoing treatment based on the condition of the patient currently undergoing treatment and an estimation model 43 generated from past treatment cases.

[0099] <Hardware configuration example> In each of the above-mentioned embodiments, each unit in the medical plan support system 10 shown in Fig. 1 or the medical plan support system 40 shown in Fig. 8 can be realized by dedicated HW (Hardware) (electronic circuitry). In Fig. 1 and Fig. 8, at least the following configurations can be considered as functional (processing) units (software modules) of a software program. Acquisition units 11 and 41, ·Classification part 12, ·Analysis Department 13, Model generation unit 14, Estimation parts 15 and 42, Outputs 16 and 44.

[0100] However, the division of the various parts shown in these drawings is for the convenience of explanation, and various configurations may be assumed when implementing the system. An example of the hardware environment in this case will be described with reference to FIG.

[0101] Fig. 9 is a diagram for explaining an example of the configuration of an information processing system 900 (computer system) capable of realizing the medical plan support system 10 according to the first embodiment of the present invention or the medical plan support system 40 according to the second embodiment. That is, Fig. 9 shows the configuration of at least one computer (information processing device) capable of realizing the medical plan support systems 10 and 40 shown in Fig. 1 and Fig. 8, and represents a hardware environment capable of realizing each function in the above-mentioned embodiments.

[0102] The information processing system 900 shown in FIG. 9 includes the following components, but may not include some of the following components. ·CPU(Central_Processing_Unit)901, ROM (Read_Only_Memory) 902, ·RAM(Random_Access_Memory)903, Hard disk (storage device) 904, A communication interface 905 for communicating with external devices; Bus 906 (communication line), A reader / writer 908 capable of reading and writing data stored in a recording medium 907 such as a CD-ROM (Compact Disc Read Only Memory), · Input / output interfaces 909 such as monitors, speakers, keyboards, etc.

[0103] That is, the information processing system 900 including the above components is a general computer in which these components are connected via a bus 906. The information processing system 900 may include a plurality of CPUs 901, or may include a CPU 901 configured with multiple cores. The information processing system 900 may include a GPU (Graphical Processing Unit) (not shown) in addition to the CPU 901.

[0104] The present invention, explained using the above-mentioned embodiment as an example, supplies a computer program capable of realizing the following functions to the information processing system 900 shown in Fig. 9. The functions are the above-mentioned configurations in the block configuration diagrams (Figs. 1 and 8) referred to in the explanation of the embodiment, or the functions of the flowcharts (Figs. 5 and 7). The present invention is then achieved by reading the computer program into the CPU 901 of the hardware, interpreting it, and executing it. The computer program supplied to the device may be stored in a readable and writable volatile memory (RAM 903), or a non-volatile storage device such as a ROM 902 or a hard disk 904.

[0105] In the above case, the method of supplying the computer program to the hardware can be a currently common procedure, such as installing the program in the device via a recording medium 907 such as a CD-ROM, or downloading the program from an external source via a communication line such as the Internet. In such a case, the present invention can be considered to be configured by the code constituting the computer program, or the recording medium 907 on which the code is stored.

[0106] The present invention has been described above by taking the above-mentioned embodiment as an exemplary example. However, the present invention is not limited to the above-mentioned embodiment. That is, the present invention can be applied in various aspects that can be understood by a person skilled in the art within the scope of the present invention.

[0107] Note that some or all of the above-described embodiments can be described as follows: However, the present invention described by way of example in the above-described embodiments is not limited to the following.

[0108] (Appendix 1) An acquisition means for acquiring patient condition information relating to a condition of a first patient; an estimation means for estimating a recommended medical action to be recommended for the first patient based on the patient condition information acquired by the acquisition means and an estimation model; an output means for outputting the recommended medical action estimated by the estimation means; Equipped with The estimation model is a model that learns a relationship between a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient. Medical planning support system.

[0109] (Appendix 2) The patient status information regarding the first patient includes a history of the biological status of the first patient and a history of medical procedures performed on the first patient. 2. A medical planning support system as described in appendix 1.

[0110] (Appendix 3) The first and second patient conditions include at least one of body temperature, heart rate, blood pressure, blood condition, urinary condition, stool condition, organ condition, consciousness state, pain state, and joint movement state; 3. The medical planning support system according to claim 1 or 2.

[0111] (Appendix 4) The recommended medical procedure and the medical procedure performed on the first and second patients include at least one of surgery, drug therapy, radiation therapy, and rehabilitation therapy; A medical planning support system according to any one of claims 1 to 3.

[0112] (Appendix 5) The surgery includes a body part to be operated on and a method of the surgery; The drug therapy includes at least one of the type, method of use, and dosage of the drug to be administered; The radiation therapy includes at least one of a type of radiation to be used, an irradiation location, and an irradiation time, The rehabilitation treatment includes at least one of a body part to be treated and a type of rehabilitation; 5. A medical planning support system as described in appendix 4.

[0113] (Appendix 6) and a model generating means for generating or updating the estimation model based on a condition of the second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient. A medical planning support system according to any one of claims 1 to 5.

[0114] (Appendix 7) The estimation means estimates a medical procedure recommended for the second patient based on the condition of the second patient, the medical procedure performed on the second patient, and the estimation model; The model generating means updating the estimation model to reduce a difference between a medical action recommended for the second patient and a medical action performed for the second patient if the change in the condition of the second patient indicates that the condition of the second patient has improved; if the change in the condition of the second patient indicates that the condition of the second patient is not improving, updating the estimation model to increase the difference between the medical action recommended for the second patient and the medical action performed for the second patient. 7. A medical planning support system as described in appendix 6.

[0115] (Appendix 8) A classification means for classifying the medical procedures performed on the second patient into groups based on their similarity and inputting the groups to the model generation means, the model generating means generates or updates the estimation model based on the group and a change in the condition of the second patient due to a medical procedure performed on the second patient; The estimation means estimates the group to which the recommended medical action belongs. 8. The medical planning support system according to claim 6 or 7.

[0116] (Appendix 9) The classification means classifies the medical procedures performed on the second patient into the groups having a hierarchical structure based on the similarity. 9. A medical planning support system as described in appendix 8.

[0117] (Appendix 10) The classification means classifies a first medical procedure using a first drug and a second medical procedure using a second drug having an equivalent effect to the first drug but of a different type into the same group. 10. The medical planning support system according to claim 8 or 9.

[0118] (Appendix 11) The first drug is a brand-name drug, and the second drug is a generic version of the first drug. 11. The medical planning support system of claim 10.

[0119] (Appendix 12) further comprising an analysis means for analyzing whether a combination of a plurality of medical procedures performed on the second patient has influenced the condition of the second patient, and inputting the combination that influenced the condition of the second patient to the model generation means; the model generating means uses the combination as a medical procedure performed on the second patient; The estimation means estimates the combination recommended for the first patient. A medical planning support system according to any one of claims 6 to 11.

[0120] (Appendix 13) the output means controls a display device to display the recommended medical action when an input operation is performed to input the patient condition information about the first patient to the estimation means. A medical planning support system according to any one of claims 1 to 12.

[0121] (Appendix 14) The output means controls the display device to display information prompting a doctor to perform an input operation indicating whether or not the recommended medical action is approved. 14. The medical planning support system of claim 13.

[0122] (Appendix 15) When the input operation by the doctor indicates approval of the recommended medical action, the output means generates control information for a medical device that performs the recommended medical action, and outputs the generated control information to the medical device. 15. The medical planning support system of claim 14.

[0123] (Appendix 16) An acquisition means for acquiring patient condition information relating to a condition of a first patient; an estimation means for estimating a recommended medical action to be recommended for the first patient based on the patient condition information acquired by the acquisition means and an estimation model; an output means for outputting the recommended medical action estimated by the estimation means; Equipped with The estimation model is a model that learns a relationship between a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient. Medical planning support device.

[0124] (Appendix 17) By the information processing device, Obtaining patient condition information relating to a condition of a first patient; Estimating a recommended medical action for the first patient based on the acquired patient condition information and an estimation model; Outputting the estimated recommended medical action; The estimation model is a model that learns a relationship between a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient. How to support medical planning.

[0125] (Appendix 18) an acquisition process for acquiring patient condition information relating to a condition of a first patient; an estimation process for estimating a recommended medical action to be recommended for the first patient based on the patient condition information acquired by the acquisition process and an estimation model; an output process for outputting the recommended medical action estimated by the estimation process; A program for causing a computer to execute the above, The estimation model is a model that learns a relationship between a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient. A recording medium on which a medical planning support program is stored. [Explanation of symbols]

[0126] 10 Medical Planning Support System 101 Patient Status Information 102 Medical Treatment Information 103 Status change information 104 Recommended Medical Procedures 11 Acquisition Department 12 Classification section 13 Analysis Department 14 Model Generation Unit 140 Estimation Model 15 Estimation part 16 Output section 20 Management terminal 200 display screen 30 Medical Equipment 40 Medical Planning Support System 400 Patient Status Information 41 Acquisition Department 42 Estimation part 420 Recommended Medical Procedures 43 Estimation model 431 Second Patient's Condition 432 Medical procedures performed on the second patient 433 Change in condition of the second patient 44 Output section 900 Information Processing Systems 901 CPU 902 ROM 903 RAM 904 Hard disk (storage device) 905 Communication Interface 906 Bus 907 Recording media 908 Reader / Writer 909 Input / Output Interface

Claims

1. acquiring means for acquiring patient condition information relating to a condition of a first patient; an estimation means for estimating a recommended medical action to be recommended for the first patient based on the patient condition information acquired by the acquisition means and an estimation model; an output means for outputting the recommended medical action estimated by the estimation means; a model generating means for generating or updating the estimation model based on a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient; A classification means for classifying the medical procedures performed on the second patient into groups based on their similarity and inputting the groups to the model generation means; Equipped with the estimation model is a model that learns, for each group, a relationship between a condition of the second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient; the model generation means generates or updates the estimation model based on a condition of the second patient, the group into which the medical procedures performed on the second patient are classified, and a change in the condition of the second patient due to the medical procedures performed on the second patient; The estimation means estimates the group to which the recommended medical action belongs. Medical planning support system.

2. The patient status information regarding the first patient includes a history of a biological status of the first patient and a history of medical procedures performed on the first patient. The medical planning support system according to claim 1 .

3. The conditions of the first patient and the second patient include at least one of body temperature, heart rate, blood pressure, blood condition, urinary condition, stool condition, organ condition, consciousness state, pain state, and joint movement state; 3. The medical planning support system according to claim 1.

4. The recommended medical procedure and the medical procedure performed on the first patient and the second patient include at least one of surgery, drug therapy, radiation therapy, and rehabilitation therapy; The medical planning support system according to any one of claims 1 to 3.

5. The surgery includes a body part to be operated on and a method of the surgery; The drug therapy includes at least one of the type, method of use, and dosage of the drug to be administered; The radiation therapy includes at least one of a type of radiation to be used, an irradiation location, and an irradiation time, The rehabilitation treatment includes at least one of a body part to be treated and a type of rehabilitation; The medical planning support system according to claim 4.

6. The estimation means estimates a medical procedure recommended for the second patient based on the condition of the second patient, the medical procedure performed on the second patient, and the estimation model; The model generating means updating the estimation model to reduce a difference between a medical action recommended for the second patient and a medical action performed for the second patient if the change in the condition of the second patient indicates that the condition of the second patient has improved; if the change in the condition of the second patient indicates that the condition of the second patient is not improving, updating the estimation model to increase the difference between a medical action recommended for the second patient and a medical action performed for the second patient. The medical planning support system according to claim 1 .

7. an analysis means for analyzing whether a combination of a plurality of medical procedures performed on the second patient has influenced the condition of the second patient, and inputting the combination that influenced the condition of the second patient to the model generation means; the model generating means uses the combination as a medical procedure performed on the second patient; The estimation means estimates the combination recommended for the first patient. The medical planning support system according to any one of claims 1 to 6.

8. the output means controls a display device to display the recommended medical action when an input operation of inputting the patient condition information related to the first patient to the estimation means is performed. The medical planning support system according to any one of claims 1 to 7.

9. By the information processing device, Obtaining patient condition information relating to a condition of a first patient; Estimating a recommended medical action for the first patient based on the acquired patient condition information and an estimation model; Outputting the estimated recommended medical action; generating or updating the estimation model based on a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient; A method for classifying medical procedures performed on the second patient into groups based on similarity thereof, and using the groups in generating or updating the estimation model, comprising: the estimation model is a model that learns, for each group, a relationship between a condition of the second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient; generating or updating the estimation model based on the condition of the second patient, the group into which the medical procedures performed on the second patient are classified, and a change in the condition of the second patient due to the medical procedures performed on the second patient; predicting the group to which the recommended medical action belongs; How to support medical planning.

10. an acquisition process for acquiring patient condition information relating to a first patient condition; an estimation process for estimating a recommended medical action to be recommended for the first patient based on the patient condition information acquired by the acquisition process and an estimation model; an output process for outputting the recommended medical action estimated by the estimation process; a model generation process for generating or updating the estimation model based on a condition of a second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient; a classification process for classifying the medical procedures performed on the second patient into groups based on their similarity and inputting the groups into the model generation process; A program for causing a computer to execute the above, the estimation model is a model that learns, for each group, a relationship between a condition of the second patient, a medical procedure performed on the second patient, and a change in the condition of the second patient due to the medical procedure performed on the second patient; the model generation process generates or updates the estimation model based on a condition of the second patient, the group into which the medical procedures performed on the second patient are classified, and a change in the condition of the second patient due to the medical procedures performed on the second patient; The estimation process estimates the group to which the recommended medical action belongs. Healthcare Planning Assistance Program.

Images