Prescription support device and prescription support method

Abstract

This prescription assistance device is provided with a control unit which: acquires attribute data indicating a patient's attribute, chronological data of the condition of the patient, and medication history data indicating a prescription history of medication for the patient; on the basis of the attribute data, the chronological data, and the medication history data, predicts the efficacy of the medication with respect to each of a plurality of prescription candidates having different combinations of the type, dose, and timing of administration of the medication; and presents a result of the prediction to a user.

Description

[Technical field]

[0001] The present disclosure relates to a prescription assistance device and a prescription assistance method. [Background technology]

[0002] Patent document 1 describes a treatment selection support system that predicts the degree of achievement of a patient's treatment goals for each treatment method, calculates the appropriateness of each treatment method for the patient, and provides information on the treatment method suitable for the patient based on the predicted degree of achievement and the calculated appropriateness. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-095960 A Summary of the Invention [Problem to be solved by the invention]

[0004] Acute heart failure is treated with drugs such as diuretics, cardiac inotropes, and infusions. To relieve systemic congestion, multiple types of diuretics are combined according to cardiac and renal function, and the diuretics, cardiac inotropes, and infusions are adjusted to prevent dehydration and maintain electrolyte balance and albumin, while ensuring urinary volume with cardiac inotropes.

[0005] The pathological conditions of patients with heart failure are extremely diverse. Although drug prescriptions are determined based on the doctor's experience, it is difficult to determine a prescription after fully understanding the patient's condition. Moreover, in cases such as heart failure treatment, where there are a wide variety of combinations of drug types, dosages, and administration timing, complex prescription decisions are required. It is difficult for conventional systems to support such complex prescription decisions.

[0006] The purpose of this disclosure is to aid in complex prescription decisions. [Means for solving the problem]

[0007] A prescription assistance device as one aspect of the present disclosure includes a control unit that acquires attribute data indicating patient attributes, time series data on the patient's condition, and medication history data indicating a drug prescription history for the patient, predicts the effects of the drugs for each of a plurality of prescription candidates that have different combinations of drug type, dosage, and administration timing based on the attribute data, the time series data, and the medication history data, and presents the prediction results to a user.

[0008] In one embodiment, the time series data includes results of vital measurements of the patient at multiple points in time, and the control unit predicts the results of the vital measurements of the patient after administration of the drug as the effect of the drug.

[0009] In one embodiment, the patient's vital signs include a urine output measurement of the patient.

[0010] In one embodiment, the time series data includes results of blood tests of the patient at multiple points in time, and the control unit predicts the results of the patient's blood tests after administration of the drug as the effect of the drug.

[0011] In one embodiment, the blood tests of the patient include at least one of a cardiac function test, a renal function test, and an electrolyte test of the patient.

[0012] In one embodiment, the control unit predicts, as the effect of the drug, the number of days until the patient is able to get out of bed after administration of the drug.

[0013] In one embodiment, the medication includes at least one of a diuretic, a vasodilator, a cardiac inotropic agent, a vasopressor, a rate regulator, an analgesic, and a fluid infusion.

[0014] In one embodiment, the control unit receives an operation to adjust at least one of the plurality of prescription candidates from the user.

[0015] In one embodiment, the control unit receives an operation from the user to select one of the plurality of prescription candidates, and controls an apparatus for administering the drug to the patient according to the selected prescription candidate.

[0016] In one embodiment, the control unit predicts the effect of the drug using a trained model that receives the attribute data, the time series data, and the medication history data as input and outputs the effect of the drug.

[0017] In one embodiment, when the drug is administered to the patient, the control unit acquires data on the patient's condition at a timing determined for each type of drug and provides feedback to the trained model.

[0018] In one embodiment, when a therapeutic intervention other than administration of the drug is performed on the patient, the control unit predicts the effect of the drug by excluding at least a portion of the data from the time series data prior to the therapeutic intervention.

[0019] In one embodiment, the intervention includes respiratory support and / or circulatory support.

[0020] A prescription support method as one aspect of the present disclosure is one in which a prescription support device acquires attribute data indicating a patient's attributes, time series data on the patient's condition, and medication history data indicating a drug prescription history for the patient, and the prescription support device predicts the effects of the drug for each of a plurality of prescription candidates having different combinations of drug type, dosage, and administration timing based on the attribute data, the time series data, and the medication history data, and the prescription support device presents the prediction results to a user. Effect of the Invention

[0021] The present disclosure can assist in making complex prescription decisions. [Brief description of the drawings]

[0022] [Figure 1] FIG. 1 is a diagram illustrating a configuration of a system according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a diagram showing the functions of a system according to one embodiment of the present disclosure and the flow of data in the system. [Diagram 3] 2 is a block diagram showing details of a prescription candidate generating function of the prescription support device according to the first embodiment. FIG. [Figure 4] 4 is a flowchart showing the operation of the prescription support device according to the first embodiment. [Diagram 5] FIG. 11 is a block diagram showing details of a prescription candidate generating function of a prescription support device according to a second embodiment. [Figure 6] 10 is a flowchart showing the operation of the prescription support device according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] The present disclosure will now be described with reference to the drawings.

[0024] In each drawing, the same or corresponding parts are denoted by the same reference numerals. In the description of each embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate.

[0025] The configuration of a system 10 according to one embodiment of the present disclosure will be described with reference to FIG.

[0026] The system 10 shown in FIG. 1 includes a prescription assistance device 20, a database 30, a pump 40, and a sensor 50.

[0027] The prescription assistance device 20 is communicatively connected to the database 30, the pump 40, and the sensor 50 directly or via a network such as a LAN. "LAN" is an abbreviation for local area network.

[0028] The prescription support device 20 is installed in a hospital, for example, but may be installed in other facilities such as a data center. The prescription support device 20 is, for example, a general-purpose computer such as a PC or a server computer, or a dedicated computer. "PC" is an abbreviation for personal computer.

[0029] The database 30 is installed in a hospital, for example, but may be installed in other facilities such as a data center. The database 30 is, for example, an RDBMS. "RDBMS" is an abbreviation for relational database management system. The database 30 is, for example, separate from the prescription support device 20, but may be integrated into the prescription support device 20.

[0030] The pump 40 is installed in a hospital. The pump 40 is, for example, an infusion pump or a syringe pump. The pump 40 may be a smart pump.

[0031] The sensor 50 is installed in a hospital. The sensor 50 is, for example, an SpO2 sensor, a heart rate sensor, a blood pressure sensor, a weight scale, a urine volume sensor, a body temperature sensor, a pulse sensor, or a respiration sensor. "SpO2" is an abbreviation for percutaneous oxygen saturation.

[0032] The configuration of the prescription support device 20 will be described with reference to FIG.

[0033] The prescription support device 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, and an output unit 25.

[0034] The control unit 21 includes at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a CPU or GPU, or a dedicated processor specialized for a particular process. "CPU" is an abbreviation for central processing unit. "GPU" is an abbreviation for graphics processing unit. The dedicated circuit is, for example, an FPGA or an ASIC. "FPGA" is an abbreviation for field-programmable gate array. "ASIC" is an abbreviation for application specific integrated circuit. The control unit 21 executes processes related to the operation of the prescription support device 20 while controlling each part of the prescription support device 20 .

[0035] The storage unit 22 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. The semiconductor memory is, for example, a RAM or a ROM. "RAM" is an abbreviation for random access memory. "ROM" is an abbreviation for read only memory. The RAM is, for example, an SRAM or a DRAM. "SRAM" is an abbreviation for static random access memory. "DRAM" is an abbreviation for dynamic random access memory. The ROM is, for example, an EEPROM. "EEPROM" is an abbreviation for electrically erasable programmable read only memory. The storage unit 22 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 22 stores data used in the operation of the prescription support device 20 and data obtained by the operation of the prescription support device 20. The database 30 may be integrated into the prescription support device 20 by being constructed in the storage unit 22.

[0036] The communication unit 23 includes at least one communication interface. The communication interface is, for example, a LAN interface. The communication unit 23 receives data used for the operation of the prescription support device 20 and transmits data obtained by the operation of the prescription support device 20.

[0037] The input unit 24 includes at least one input interface. The input interface is, for example, a physical key, a capacitive key, a pointing device, a touch screen provided integrally with a display, or a microphone. The input unit 24 receives an operation for inputting data used for the operation of the prescription support device 20. Instead of being provided in the prescription support device 20, the input unit 24 may be connected to the prescription support device 20 as an external input device. As the connection method, for example, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used. "USB" is an abbreviation for Universal Serial Bus. "HDMI (registered trademark)" is an abbreviation for High-Definition Multimedia Interface.

[0038] The output unit 25 includes at least one output interface. The output interface is, for example, a display or a speaker. The display is, for example, an LCD or an organic EL display. "LCD" is an abbreviation for liquid crystal display. "EL" is an abbreviation for electro luminescence. The output unit 25 outputs data obtained by the operation of the prescription support device 20. Instead of being provided in the prescription support device 20, the output unit 25 may be connected to the prescription support device 20 as an external output device. As the connection method, for example, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used.

[0039] The functions of the prescription support device 20 are realized by executing a program as one aspect of the present disclosure in a processor corresponding to the control unit 21. That is, the functions of the prescription support device 20 are realized by software. The program causes a computer to execute the operations of the prescription support device 20, thereby causing the computer to function as the prescription support device 20. That is, the computer functions as the prescription support device 20 by executing the operations of the prescription support device 20 in accordance with the program.

[0040] The program may be stored in a non-transitory computer-readable medium. Examples of the non-transitory computer-readable medium include a flash memory, a magnetic recording device, an optical disk, a magneto-optical recording medium, or a ROM. The program may be distributed, for example, by selling, transferring, or lending a portable medium such as an SD card, DVD, or CD-ROM on which the program is stored. "SD" is an abbreviation for Secure Digital. "DVD" is an abbreviation for digital versatile disc. "CD-ROM" is an abbreviation for compact disc read only memory. The program may be distributed by storing the program in the storage of a server and transferring the program from the server to another computer. The program may be provided as a program product.

[0041] A computer temporarily stores a program stored in a portable medium or transferred from a server in a main storage device. Then, the computer reads the program stored in the main storage device with a processor and executes processing according to the read program with the processor. The computer may read the program directly from a portable medium and execute processing according to the program. The computer may execute processing according to the received program each time a program is transferred from the server to the computer. Processing may be executed by a so-called ASP-type service that realizes functions only by execution instructions and result acquisition without transferring a program from the server to the computer. "ASP" is an abbreviation for application service provider. Programs include information used for processing by a computer and equivalent to a program. For example, data that is not a direct command to a computer but has the property of defining computer processing falls under " equivalent to a program.

[0042] A part or all of the functions of the prescription support device 20 may be realized by a dedicated circuit corresponding to the control unit 21. In other words, a part or all of the functions of the prescription support device 20 may be realized by hardware.

[0043] The function of the system 10 according to one embodiment of the present disclosure and the flow of data in the system 10 will be described with reference to FIG.

[0044] The prescription support device 20 has a prescription candidate generation function 61 and an effect prediction function 62.

[0045] In the database 30, attribute data 71 indicating the attributes of a patient, time-series data 72 on the condition of the patient, and medication history data 73 indicating the prescription history of medicines for the patient are registered.

[0046] The attribute data 71 is input from another system or manually input. The attribute data 71 includes data indicating at least one of the following as attributes of the patient: age, sex, blood type, and comorbidity.

[0047] The time series data 72 includes sensor data 74, test data 75, and manually input data 76. The sensor data 74 is input from the sensor 50. The sensor data 74 includes data indicating at least one of, for example, SpO2, heart rate, cardiac output, cardiac index, blood pressure, weight, left ventricular end diastolic pressure, left ventricular ejection fraction, central venous pressure, urine volume, respiratory rate, pulmonary artery wedge pressure, and total body water content as the patient's condition. The test data 75 is input from another system or manually input. The test data 75 includes data indicating at least one of, for example, liver function, kidney function, creatinine, eGFR, BNP, NT-proBNP, hemoglobin, sodium, potassium, BUN, albumin, and CRP as the patient's condition. "eGFR" is an abbreviation for estimated glomerular filtration rate. "BNP" is an abbreviation for brain natriuretic peptide. "NT-proBNP" is an abbreviation for N-terminal fragment of pro-brain natriuretic peptide. "BUN" is an abbreviation for blood urea nitrogen. "CRP" is an abbreviation for C-reactive protein. The manually input data 76 is manually input. The manually input data 76 includes data indicating at least one of the following conditions of the patient: presence or absence of edema such as lower leg edema, presence or absence of coldness in the extremities, presence or absence of cyanosis, presence or absence of jugular vein distension, presence or absence of orthopnea, urine volume, and body weight.

[0048] The medical history data 73 is input from another system or manually input. The medical history data 73 includes data indicating at least one of the drug type, dosage, administration method, administration start date and time, and administration end date and time.

[0049] An electronic medical record 77 is further registered in the database 30. At least one of the attribute data 71, the time-series data 72, and the medical history data 73 may be registered as a part of the electronic medical record 77.

[0050] The prescription support device 20 acquires attribute data 71, time-series data 72, medical history data 73, and an electronic medical record 77 from the database 30. The prescription support device 20 may acquire sensor data 74 included in the time-series data 72 directly from the sensor 50.

[0051] The control unit 21 of the prescription support device 20 generates a plurality of prescription candidates with different combinations of drug type, dosage, and administration timing using the prescription candidate generation function 61. The control unit 21 predicts the drug effect for each of the generated plurality of prescription candidates based on the attribute data 71, the time series data 72, and the medication history data 73 using the effect prediction function 62. The control unit 21 presents the calculation result 78, which is the result of the prediction, to a user such as a doctor.

[0052] According to the above-mentioned aspect, it is possible to support the judgment of a complicated prescription. That is, even in the case of a wide variety of combinations of drug types, dosages, and administration timings, such as in the treatment of heart failure, it is possible to inform the user of the effects of each combination when applied to a patient, making it easier to decide on a prescription.

[0053] The prescription candidate generating function 61 may be omitted. For example, the control unit 21 of the prescription support device 20 may receive input of multiple prescription candidates from a user via the input unit 24 such as a touch screen, instead of generating multiple prescription candidates. Alternatively, the control unit 21 may receive multiple prescription candidates devised by other doctors and shared in an external system such as a cloud computing system, via the communication unit 23. Alternatively, the control unit 21 may read multiple prescription candidates previously stored in the storage unit 22.

[0054] Hereinafter, several embodiments will be described with reference to the drawings as specific examples of the aspects shown in FIGS.

[0055] (First embodiment) In this embodiment, as shown in FIG. 3, the prescription candidate generation function 61 of the prescription support device 20 includes a prescription range setting function 63, a prescription range memory function 64, a previous prescription input function 65, a prescription candidate creation function 66, and a prescription candidate output function 67.

[0056] The operation of the prescription support device 20 according to this embodiment will be described with reference to Fig. 4. This operation corresponds to the prescription support method according to this embodiment.

[0057] In step S101, the control unit 21 acquires the attribute data 71.

[0058] In this embodiment, the control unit 21 accepts input of the attribute data 71 via the database 30. Specifically, the control unit 21 receives the attribute data 71 input to the database 30 from the database 30 via the communication unit 23.

[0059] As a modification of this embodiment, the control unit 21 may receive input of the attribute data 71 from the user via the input unit 24 such as a touch screen.

[0060] In step S102, the control unit 21 acquires the time-series data 72.

[0061] In this embodiment, the control unit 21 accepts input of time-series data 72 via the database 30. Specifically, the control unit 21 receives, from the database 30, the sensor data 74, the inspection data 75, and the manually input data 76 input to the database 30 via the communication unit 23.

[0062] As a modified example of this embodiment, the control unit 21 may directly receive the input of the sensor data 74 from the sensor 50. Specifically, the control unit 21 may receive the sensor data 74 obtained by the sensor 50 directly from the sensor 50 via the communication unit 23. Alternatively, the control unit 21 may receive the input of the manually input data 76 and other data included in the time-series data 72 from the user via the input unit 24 such as a touch screen.

[0063] In step S103, the control unit 21 acquires the medication history data 73.

[0064] In this embodiment, the control unit 21 accepts input of the medication history data 73 via the database 30. Specifically, the control unit 21 receives the medication history data 73 input to the database 30 from the database 30 via the communication unit 23.

[0065] As a modification of this embodiment, the control unit 21 may accept input of the medication history data 73 from the user via the input unit 24 such as a touch screen.

[0066] In step S104, the control unit 21 generates a plurality of prescription candidates using the prescription candidate generating function 61.

[0067] Specifically, the control unit 21 sets the prescription range as a prescription range, which is a limited range of the type, dosage, and administration timing of a drug that may be prescribed, using a prescription range setting function 63. The control unit 21 stores the set prescription range in the storage unit 22 using a prescription range storage function 64. The control unit 21 accepts input of the prescription previously applied to the patient from the user via the input unit 24 such as a touch screen using a previous prescription input function 65. The control unit 21 creates multiple prescription candidates by combining multiple patterns of drug type, dosage, and administration timing based on the previous prescription input using a prescription candidate creation function 66 so that the prescription falls within the prescription range stored in the storage unit 22. The control unit 21 outputs the created multiple prescription candidates using a prescription candidate output function 67.

[0068] In this embodiment, when creating multiple prescription candidates, the control unit 21 refers to the attribute data 71 acquired in step S101 and adjusts the multiple prescription candidates to be created according to the patient's attributes, but the control unit 21 may create multiple prescription candidates without referring to the attribute data 71.

[0069] In this embodiment, when creating multiple prescription candidates, the control unit 21 refers to the time series data 72 acquired in step S102 and adjusts the multiple prescription candidates to be created according to the patient's condition, but the multiple prescription candidates may be created without referring to the time series data 72.

[0070] In this embodiment, when creating multiple prescription candidates, the control unit 21 refers to the medication history data 73 acquired in step S103 and adjusts the multiple prescription candidates to be created according to the drug prescription history for the patient, but multiple prescription candidates may be created without referring to the medication history data 73.

[0071] As an example, the prescribed medication is a medication used to treat heart failure, and specifically includes at least one of a diuretic, a vasodilator, a cardiac inotropic, a vasopressor, a heart rate regulator, an analgesic, and a fluid infusion.

[0072] In step S105, the control unit 21 predicts, using the effect prediction function 62, the effect of the drug for each of the multiple prescription candidates generated in step S104 based on the attribute data 71 acquired in step S101, the time series data 72 acquired in step S102, and the medication history data 73 acquired in step S103.

[0073] Specifically, the control unit 21 predicts the effect of the drug using a trained model that receives attribute data 71, time-series data 72, and medication history data 73 as inputs and outputs the effect of the drug. The trained model may be constructed by any method, but in this embodiment, it is constructed by machine learning. As the machine learning method, any method such as RNN, SVM, or random forest can be used. "RNN" is an abbreviation for recurrent neural network. "SVM" is an abbreviation for support-vector machine.

[0074] In this embodiment, the time series data 72 includes the results of vital measurements of the patient at multiple points in time. The control unit 21 predicts the results of the vital measurements of the patient after administration of the drug as the effect of the drug. As an example, the vital measurements of the patient include a urine volume measurement of the patient.

[0075] As an example, the results of vital sign measurement of a patient include values ​​related to cardiac failure treatment, such as SpO2, heart rate, cardiac output, cardiac index, blood pressure, weight, left ventricular end-diastolic pressure, left ventricular ejection fraction, central venous pressure, urine volume, respiratory rate, pulmonary artery wedge pressure, total body water, liver function, kidney function, creatinine, eGFR, BNP, NT-proBNP, hemoglobin, sodium, potassium, BUN, albumin, CRP, presence or absence of edema such as lower leg edema, presence or absence of cold limbs, presence or absence of cyanosis, presence or absence of jugular vein distension, and presence or absence of orthopnea.

[0076] In this embodiment, the time-series data 72 includes the results of the patient's blood test at multiple points in time. The control unit 21 predicts the results of the patient's blood test after administration of the drug as the effect of the drug. As an example, the patient's blood test includes at least one of the patient's cardiac function test, renal function test, and electrolyte test.

[0077] The control unit 21 may predict, as the effect of the drug, the number of days until the patient is able to get out of bed after administration of the drug.

[0078] In step S106, the control unit 21 presents the calculation result 78, which is the result of the prediction in step S105, to the user.

[0079] Specifically, the control unit 21 causes the calculation result 78 to be output to the output unit 25 such as a display.

[0080] As an example, the effect of medication on renal function for an acute heart failure patient is predicted. In this case, in step S102, sensor data 74 including vital measurements such as blood pressure, heart rate, and urinary flow rate, and test data 75 including blood test values ​​such as renal function, liver function, and albumin are acquired. In step S105, urine flow rate and creatinine or eGFR are predicted as the effects of the medication. Sodium, chloride, potassium, blood pressure, heart rate, cardiac output, left ventricular end-diastolic pressure, left ventricular ejection fraction, and central venous pressure may also be predicted as the effects of the medication.

[0081] For example, in the treatment of heart failure, if the timing of checking vital signs and urine volume to decide on a prescription is delayed due to the doctor's work schedule, the timing of switching prescriptions will be delayed and the organ congestion state may be prolonged. According to the above example, a quick decision can be made and the organ congestion state can be quickly ended. The task of checking vital signs and urine volume is no longer necessary, which can also reduce the burden on the doctor.

[0082] For example, a patient's renal function is reduced and the patient is mainly suffering from fluid overload. For patients who cannot urinate with a normal amount of diuretic, increasing the diuretic dose can quickly ensure sufficient urine volume, but a large increase in the dose can cause dehydration due to rapid water removal, which may lead to further decline in renal function. In some cases, it may be possible to ensure sufficient urine volume by combining with other diuretics, but in this case, electrolyte abnormalities may occur. In this embodiment, it is possible to predict the progression of renal function, electrolytes, vital signs, etc. for each prescription candidate, making it easier to determine the appropriate amount of drug and combination of drugs.

[0083] As a specific example, in patients with reduced renal function, furosemide is usually intravenously injected at a small amount of 5 mg to 10 mg, and the amount is increased while checking the reaction, and continuous administration is also considered, but by applying this embodiment, it becomes possible to consider what administration method should be used to administer an appropriate amount of furosemide from the early stage of treatment.By applying this embodiment, it becomes possible to quickly release congestion while maintaining renal function and electrolyte balance.

[0084] As described above, in this embodiment, the control unit 21 of the prescription support device 20 accepts input of the patient's time series data 72. The control unit 21 accepts input of attribute data 71, which is basic data of the patient. The control unit 21 accepts input of medication history data 73, which indicates the prescription history of medicines. The control unit 21 generates multiple prescription candidates from the time series data 72, the attribute data 71, and the medication history data 73. The control unit 21 infers the patient's outcome for each prescription candidate from the time series data 72, the attribute data 71, and the medication history data 73. The outcome is specifically the effect of the medicine. As an example, the outcome is a parameter related to the patient's cardiac function and renal function, and a parameter related to complications accompanying the treatment. The control unit 21 outputs the inferred result of the outcome.

[0085] In this embodiment, by selecting a complex prescription combination from several combinations, it is possible to simultaneously simplify the prescription and improve the outcome.

[0086] As a modified example of this embodiment, the control unit 21 may receive an operation to adjust at least one of the multiple prescription candidates from the user. That is, the multiple prescription candidates may be changed by the user.

[0087] As a modified example of this embodiment, when a drug is administered to a patient, the control unit 21 may acquire data on the patient's condition at a timing determined for each type of drug, and provide feedback to the trained model used in step S105.

[0088] As a modified example of this embodiment, when a therapeutic intervention other than administration of a drug is performed on a patient, the control unit 21 may predict the effect of the drug by excluding at least a part of data from the time-series data 72 that precedes the therapeutic intervention. As an example, the therapeutic intervention includes at least one of respiratory support and blood circulation support. The respiratory support is performed, for example, using an artificial ventilator. The blood circulation support is performed, for example, as a surgery such as a coronary artery bypass surgery.

[0089] As a modified example of this embodiment, the control unit 21 may generate a plurality of prescription candidates using a trained model that receives the attribute data 71, the time-series data 72, and the medication history data 73 as inputs and outputs a plurality of prescription candidates. The previous prescription may be included in the input of the trained model.

[0090] Second embodiment The differences from the first embodiment will be mainly described.

[0091] In this embodiment, as shown in FIG. 5, the prescription candidate generating function 61 of the prescription support device 20 includes a drug combination setting function 68 and a drug combination storage function 69 instead of the previous prescription input function 65.

[0092] The operation of the prescription support device 20 according to this embodiment will be described with reference to Fig. 6. This operation corresponds to the prescription support method according to this embodiment.

[0093] The processing from step S201 to step S203 is the same as the processing from step S101 to step S103 in the first embodiment, and therefore the description thereof will be omitted.

[0094] In step S204, the control unit 21 ends the process when an operation instructing the end of the process is received from the user via the input unit 24 such as a touch screen. If such an operation is not performed, the process of step S205 is performed.

[0095] In step S205, the control unit 21 generates a plurality of prescription candidates using the prescription candidate generating function 61.

[0096] Specifically, the control unit 21 sets the prescription range as a prescription range, which is a limited range of the type, dosage, and administration timing of a drug that may be prescribed, using the prescription range setting function 63. The control unit 21 stores the set prescription range in the storage unit 22 using the prescription range storage function 64. The control unit 21 accepts input of one or more combinations of the type, dosage, and administration timing of a drug from the user via the input unit 24 such as a touch screen using the drug combination setting function 68. The control unit 21 stores the input combination in the storage unit 22 as a user-specified drug combination using the drug combination storage function 69. The control unit 21 creates multiple prescription candidates by combining the type, dosage, and administration timing of a drug in multiple patterns based on the user-specified drug combination stored in the storage unit 22 using the prescription candidate creation function 66 so that the drug combination falls within the prescription range stored in the storage unit 22. If the user-specified drug combination falls within the prescription range, the control unit 21 may adopt the user-specified drug combination as a prescription candidate as it is. The control unit 21 outputs the created multiple prescription candidates using the prescription candidate output function 67.

[0097] The processes in steps S206 and S207 are the same as those in steps S105 and S106 in the first embodiment, and therefore will not be described.

[0098] In step S208, the control unit 21 accepts an operation from the user to select one of the multiple prescription candidates output in step S205.

[0099] Specifically, the control unit 21 receives an operation from the user via an input unit 24 such as a touch screen to select one of a plurality of prescription candidates outputted to an output unit 25 such as a display.

[0100] In steps S209 and S210, the control unit 21 controls a device for administering a drug to a patient in accordance with the prescription candidate selected in step S208.

[0101] Specifically, the control unit 21 controls a pump 40, such as a smart pump, as a device for administering a drug to a patient so that the type of drug specified in a prescription candidate selected by the user is administered to the patient in the dosage amount specified in the prescription candidate at the administration timing specified in the prescription candidate.

[0102] In step S211, the control unit 21 presents the result of the drug administration in step S210 to the user.

[0103] Specifically, the control unit 21 causes the output unit 25, such as a display, to output the results of the drug administration.

[0104] After step S211, the processes of steps S202, S203, and S204 are performed again.

[0105] As described above, in this embodiment, the user can select a prescription according to the outcome inference result. Furthermore, the prescribed drug can be delivered to the patient according to the prescription selection result of the user. The prescription may be output to the electronic medical record 77 according to the prescription selection result.

[0106] As a modified example of this embodiment, the timing for the control unit 21 to acquire the time series data 72 in step S202 following step S210 may be set according to the prescription applied in step S210. This setting may be performed manually by the user or automatically by the control unit 21. Furthermore, the set timing may be output as time on the output unit 25 such as a display. If the time series data 72 is not acquired at that time or within a certain period of time from that time, a warning may be output on the output unit 25 such as a display.

[0107] As a specific example, when a loop diuretic such as Lasix (registered trademark) is prescribed, the urine volume 15 to 20 minutes and 2 to 4 hours later could be predicted, and actual urine volume data could be obtained 15 to 20 minutes and 2 to 4 hours later and fed back.

[0108] As another specific example, it is conceivable to predict the urine volume five minutes later at the time a vasopressor such as dobutamine is prescribed, and then obtain the actual urine volume data five minutes later and feed it back.

[0109] The present disclosure is not limited to the above-described embodiments. For example, multiple blocks shown in the block diagram may be integrated, or one block may be divided. Instead of executing multiple steps shown in the flowchart in chronological order as described, each step may be executed in parallel or in a different order depending on the processing capacity of the device executing each step, or as necessary. Other modifications are possible without departing from the spirit of the present disclosure. [Explanation of symbols]

[0110] 10. System 20 Prescription Support Device 21 Control section 22 Memory section 23 Communications Department 24 Input section 25 Output section 30 Databases 40 Pump 50 Sensors 61 Prescription candidate generation function 62 Effect prediction function 63 Prescription range setting function 64 Prescription range memory function 65 Previous prescription input function 66 Prescription candidate creation function 67 Prescription candidate output function 68 Drug combination setting function 69 Drug combination memory function 71 Attribute Data 72 Time Series Data 73 Medical History Data 74 Sensor Data 75 Test Data 76 Manually entered data 77 Electronic Medical Records 78 Calculation results

Claims

1. a control unit that acquires attribute data indicating attributes of a patient, time-series data on the condition of the patient, and medication history data indicating a prescription history of drugs for the patient, predicts effects of the drugs for each of a plurality of prescription candidates having different combinations of drug type, dosage, and administration timing based on the attribute data, the time-series data, and the medication history data, and presents the prediction results to a user; The control unit is a prescription support device that, when a therapeutic intervention other than administration of the drug is performed on the patient, predicts the effectiveness of the drug by excluding at least a portion of the data from the time series data that precedes the therapeutic intervention.

2. the time series data includes vital signs of the patient at a plurality of time points; The prescription support device according to claim 1 , wherein the control unit predicts, as the effect of the drug, a result of vital sign measurement of the patient after administration of the drug.

3. the time series data includes results of blood tests of the patient at multiple time points; The prescription support device according to claim 1 , wherein the control unit predicts, as the effect of the drug, a result of a blood test of the patient after administration of the drug.

4. The prescription support device according to claim 1 , wherein the control unit predicts, as the effect of the drug, the number of days until the patient is able to get out of bed after administration of the drug.

5. The prescription support device according to claim 1 , wherein the control unit receives an operation to adjust at least one of the plurality of prescription candidates from the user.

6. The prescription support device according to any one of claims 1 to 5, wherein the control unit receives an operation from the user to select one of the plurality of prescription candidates, and controls an apparatus for administering the drug to the patient according to the selected prescription candidate.

7. The prescription support device according to any one of claims 1 to 6, wherein the control unit predicts the effect of the drug using a trained model that receives the attribute data, the time series data, and the medication history data as inputs and outputs the effect of the drug.

8. The prescription support device according to claim 7 , wherein the control unit, when the drug is administered to the patient, acquires data on the patient's condition at a timing determined for each type of drug and provides feedback to the trained model.

9. The prescription support device according to claim 1 , wherein the therapeutic intervention includes at least one of respiratory support and blood circulation support.

10. A memory unit that stores a prescription range including a limit range of a drug type, a dosage, and a timing of administration; a control unit that acquires attribute data indicating attributes of a patient, time-series data on the condition of the patient, and medication history data indicating a prescription history of the drug for the patient, creates a plurality of prescription candidates with different combinations of the drug type, dosage, and administration timing by combining the drug type, dosage, and administration timing in a plurality of patterns so as to fall within the prescription range stored in the storage unit, predicts the effect of the drug for each of the plurality of prescription candidates based on the attribute data, the time-series data, and the medication history data, and presents the prediction result to a user; A prescription assistance device comprising:

11. Further comprising an input unit, The control unit receives input of one or more combinations of the drug type, dosage, and administration timing from the user via the input unit, stores the input one or more combinations in the memory unit as a user-specified drug combination, and creates the multiple prescription candidates by combining the drug type, dosage, and administration timing in multiple patterns based on the user-specified drug combination stored in the memory unit so that the combination falls within the prescription range stored in the memory unit.

12. Further comprising an input unit, The prescription support device according to claim 10, wherein the control unit receives input of the prescription previously applied to the patient from the user via the input unit, and creates the multiple prescription candidates by combining the type, dosage, and administration timing of the drug in multiple patterns based on the input prescription so that the combination falls within the prescription range stored in the memory unit.

13. Further comprising an input unit and an output unit, 11. The prescription support device according to claim 10, wherein the control unit outputs the plurality of prescription candidates to the user via the output unit, receives an operation from the user to select one of the plurality of prescription candidates via the input unit, and controls an equipment that administers the drug to the patient so that the type of drug specified in the selected prescription candidate is administered to the patient in the dosage specified in the selected prescription candidate at the administration timing specified in the selected prescription candidate.

14. A prescription support device acquires attribute data indicating attributes of a patient, time-series data on a condition of the patient, and medication history data indicating a prescription history of a drug for the patient; The prescription support device predicts an effect of the drug for each of a plurality of prescription candidates having different combinations of the drug type, dosage, and administration timing based on the attribute data, the time-series data, and the medication history data; A prescription support method in which the prescription support device presents a result of the prediction to a user, The prescription support method includes a prescription support device that, when a therapeutic intervention other than administration of the drug is performed on the patient, predicts the effectiveness of the drug by excluding at least a portion of the data from the time series data that precedes the therapeutic intervention.

Images