Biological function estimation device

Abstract

This biological function estimation device comprises a control unit for predicting drug reaction of a patient on the basis of data of a drug administered to the patient and an evaluation value obtained by evaluating a biological function of the patient from an examination result prior to drug administration, and correcting the evaluation value in accordance with the result of comparing the prediction value and an actual measured value obtained by actual measurement of the drug reaction after drug administration.

Description

[Technical field]

[0001] The present disclosure relates to a biological function estimation device and a biological function estimation method. [Background technology]

[0002] Patent Document 1 describes a drug administration system that measures and analyzes a patient's blood values ​​and calculates drug parameters. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2013-520718 Summary of the Invention [Problem to be solved by the invention]

[0004] Accurate and real-time evaluation of renal function in hospitalized patients is important for improving the efficacy of treatment and for suppressing the occurrence of side effects such as renal dysfunction. However, conventional methods have not achieved sufficient evaluation of renal function due to the need for blood tests each time or problems with the characteristics or reliability of the indices used for evaluation.

[0005] An object of the present disclosure is to evaluate a patient's biological functions with high accuracy without the need for testing each time. [Means for solving the problem]

[0006] A biological function estimation device as one aspect of the present disclosure is provided with a control unit that predicts a drug response of a patient based on an evaluation value obtained by evaluating the patient's biological function from test results before drug administration and data on the drug to be administered to the patient, and corrects the evaluation value depending on a comparison result between the obtained predicted value and an actual measured value obtained by actually measuring the drug response after drug administration.

[0007] In one embodiment, the control unit changes at least one of a formula and a coefficient for calculating the evaluation value according to a result of the comparison.

[0008] In one embodiment, the data includes at least one of the following: type of drug, total amount administered, and amount administered per unit time.

[0009] In one embodiment, the control unit predicts the drug response for drug administration at each of multiple time points, and corrects the evaluation value based on the comparison result between the new predicted value obtained and a new actual value obtained by actually measuring the drug response after drug administration.

[0010] In one embodiment, the control unit predicts the drug response for drug administration at a second time point after the first time point among the multiple time points based on the corrected evaluation value obtained for drug administration at the first time point and data on the drug administered to the patient at the second time point, and further corrects the evaluation value depending on the comparison result between the obtained new predicted value and a new actual value obtained by actually measuring the drug response after drug administration.

[0011] In one embodiment, the control unit records the corrected evaluation value obtained for drug administration at each of the multiple time points as a provisional correction value, and further corrects the evaluation value using the recorded multiple provisional correction values.

[0012] In one embodiment, the control unit sets a pattern of drug administration at each of the multiple time points.

[0013] In one embodiment, the control unit adjusts the pattern at the next or subsequent time points depending on a comparison result between a new predicted value obtained for drug administration at any one of the multiple time points and a new actual value.

[0014] In one embodiment, the control unit selects, as the pattern, a different pattern candidate for each time point from a predetermined pattern candidate group.

[0015] In one embodiment, the control unit sets the pattern by setting at least one of the number of times a drug is administered and the dosage per unit time.

[0016] In one embodiment, the control unit sets the pattern by setting a timing for administering a drug.

[0017] In one embodiment, when setting the pattern, the control unit refers to performance data indicating a history of past drug administrations and at least one of predicted values, actual measured values, and corrected values ​​of evaluation values.

[0018] In one embodiment, the history shown in the performance data includes the history of other patients.

[0019] In one embodiment, the control unit adjusts the timing of measuring the drug response according to the pattern.

[0020] In one embodiment, the control unit refers to historical data indicating a history of past drug administrations and drug reactions when predicting the drug response.

[0021] In one embodiment, the history shown in the progress data includes the history of other patients.

[0022] In one embodiment, the control unit controls a pump that administers a drug to the patient.

[0023] In one embodiment, the control unit controls a sensor that measures the drug response.

[0024] In one embodiment, the biological function estimation apparatus further includes an output unit that outputs a corrected value of the evaluation value.

[0025] In one embodiment, the control unit acquires the inspection result and calculates the evaluation value from the inspection result.

[0026] In one embodiment, the evaluation value is a value obtained by evaluating renal function as the biological function, and the control unit predicts a urine volume as the drug response.

[0027] A biological function estimation method as one aspect of the present disclosure involves using a computer to predict a drug response of a patient based on an evaluation value obtained by evaluating the patient's biological function from test results before drug administration and data on the drug to be administered to the patient, administering the drug to the patient using a pump, measuring the drug response after drug administration using a sensor, and correcting the evaluation value based on a comparison result between the predicted value and the actual measured value obtained by the computer. Effect of the Invention

[0028] According to the present disclosure, it is possible to evaluate a patient's biological functions with high accuracy without the need for testing each time. [Brief description of the drawings]

[0029] [Figure 1] FIG. 1 is a diagram illustrating the configuration and functions of a system according to one embodiment of the present disclosure. [Diagram 2] 1 is a block diagram showing a configuration of a biological function estimation device according to an embodiment of the present disclosure. [Diagram 3] 1 is a diagram illustrating the configuration and functions of a system according to a first embodiment. [Figure 4] 4 is a flowchart showing the operation of the biological function estimation apparatus according to the first embodiment. [Diagram 5] FIG. 11 is a diagram illustrating the configuration and functions of a system according to a second embodiment. [Figure 6] 10 is a flowchart showing the operation of the biological function estimation apparatus according to the second embodiment. [Figure 7] FIG. 11 is a diagram illustrating the configuration and functions of a system according to a third embodiment. [Figure 8] 13 is a flowchart showing the operation of the biological function estimation apparatus according to the third embodiment. [Figure 9]FIG. 13 is a diagram showing the configuration and functions of a system according to a fourth embodiment. [Figure 10] 13 is a flowchart showing the operation of the biological function estimation apparatus according to the fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] The present disclosure will now be described with reference to the drawings.

[0031] 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.

[0032] The configuration of a system 10 according to one embodiment of the present disclosure will be described with reference to FIG.

[0033] The system 10 shown in FIG. 1 includes a biological function estimation device 20, a database 30, a pump 40, and a sensor 50.

[0034] The biological function estimation device 20 is communicatively connected to a database 30, a pump 40, and a sensor 50 directly or via a network such as a LAN. "LAN" is an abbreviation for local area network.

[0035] The biological function estimation apparatus 20 is installed in a hospital, for example, but may be installed in other facilities such as a data center. The biological function estimation apparatus 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.

[0036] 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 biological function estimation device 20, but may be integrated into the biological function estimation device 20.

[0037] 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.

[0038] The sensor 50 is installed in a hospital. The sensor 50 is, for example, a urine volume sensor, a body temperature sensor, a blood pressure sensor, a pulse sensor, or a respiration sensor.

[0039] The configuration of a biological function estimation device 20 according to one embodiment of the present disclosure will be described with reference to FIG.

[0040] The biological function estimation device 20 includes a control unit 21 , a storage unit 22 , a communication unit 23 , an input unit 24 , and an output unit 25 .

[0041] The control unit 21 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a CPU or GPU, or a dedicated processor specialized for specific processing. "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 processing related to the operation of the biological function estimation device 20 while controlling each part of the biological function estimation device 20.

[0042] 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 biological function estimation device 20 and data obtained by the operation of the biological function estimation device 20. The database 30 may be integrated into the biological function estimation device 20 by being constructed in the storage unit 22.

[0043] 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 in the operation of the biological function estimation device 20, and transmits data obtained by the operation of the biological function estimation device 20.

[0044] 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 integral with a display, or a microphone. The input unit 24 accepts an operation to input data used for the operation of the biological function estimation device 20. The input unit 24 may be connected to the biological function estimation device 20 as an external input device instead of being provided in the biological function estimation device 20. As a connection method, 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.

[0045] 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 biological function estimation device 20. The output unit 25 may be connected to the biological function estimation device 20 as an external output device instead of being provided in the biological function estimation device 20. As a connection method, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used.

[0046] The functions of the biological function estimation apparatus 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 biological function estimation apparatus 20 are realized by software. The program causes a computer to execute the operations of the biological function estimation apparatus 20, thereby causing the computer to function as the biological function estimation apparatus 20. That is, the computer functions as the biological function estimation apparatus 20 by executing the operations of the biological function estimation apparatus 20 in accordance with the program.

[0047] The program may be stored in a non-transitory computer-readable medium. Examples of the non-transitory computer-readable medium include 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 a DVD or CD-ROM on which the program is stored. "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.

[0048] 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.

[0049] A part or all of the functions of the biological function estimation device 20 may be realized by a dedicated circuit equivalent to the control unit 21. In other words, a part or all of the functions of the biological function estimation device 20 may be realized by hardware.

[0050] 1 and 2, a function of the system 10 according to one embodiment of the present disclosure will be described. This function corresponds to a biological function estimation method according to one embodiment of the present disclosure.

[0051] The database 30 holds in-hospital data such as device measurements, test data, and electronic medical record data. The test data includes test results before drug administration.

[0052] The control unit 21 of the biological function estimation device 20 acquires the in-hospital data from the database 30 via the communication unit 23. The control unit 21 evaluates the biological function of the patient from the test results before drug administration, which are included in the in-hospital data. As a result, an evaluation value Ve is obtained. The biological function is, for example, an internal organ function such as renal function, cardiac function, pulmonary function, or liver function. The evaluation of the biological function may use not only the test results before drug administration, but also device measurement values ​​and electronic medical record data before drug administration, which are included in the in-hospital data. As a modified example, the biological function of the patient may be evaluated by a device other than the biological function estimation device 20. That is, the control unit 21 of the biological function estimation device 20 may acquire the evaluation value Ve from another device that evaluates the biological function of the patient from the test results before drug administration, via the communication unit 23.

[0053] The control unit 21 of the biological function estimation device 20 acquires data Dm of a drug to be administered to a patient from the pump 40 or a device that monitors and controls the pump 40, via the communication unit 23. The data Dm includes at least one of the type of drug, the total administered amount, and the administered amount per unit time.

[0054] The control unit 21 of the biological function estimation device 20 predicts the drug response of the patient based on the evaluation value Ve and the drug data Dm. As a result, a predicted value Vp is obtained. The drug response is, for example, a vital sign such as urine volume, body temperature, blood pressure, pulse rate, or respiration, or a change in the vital sign.

[0055] The pump 40 administers a drug to the patient. The sensor 50 measures the drug response after the drug is administered, resulting in a measured value Vm.

[0056] The control unit 21 of the biological function estimation device 20 acquires the actual measurement value Vm from the sensor 50 via the communication unit 23.

[0057] The control unit 21 of the biological function estimation device 20 corrects the evaluation value Ve according to the result of comparison between the predicted value Vp and the actual measurement value Vm.

[0058] According to the above-described aspect, it is possible to evaluate the biological functions of a patient with high accuracy without the need for testing each time.

[0059] Hereinafter, several embodiments will be described with reference to the drawings as specific examples of the aspects shown in FIGS.

[0060] (First embodiment) In this embodiment, as shown in FIG. 3, the in-hospital data held by the database 30 includes blood test values ​​as test results before drug administration. The blood test values ​​included in the in-hospital data are, for example, Cre, BUN / Cre, eGFR, Alb, or an inflammatory marker. "Cre" is the creatinine value. "BUN / Cre" is the value obtained by dividing the concentration of urea nitrogen contained in the blood by the concentration of creatinine. "eGFR" is the estimated glomerular filtration rate. "Alb" is the albumin value. The blood test value included in the in-hospital data may be Cys-C. "Cys-C" is the serum cystatin C value. The electronic medical record data included in the in-hospital data is, for example, blood pressure or pulse rate.

[0061] In this embodiment, the evaluation value Ve is a value obtained by evaluating renal function as a biological function. The evaluation value Ve is, for example, GFR or eGFR. "GFR" is glomerular filtration rate. The control unit 21 of the biological function estimation device 20 predicts the urine volume per unit time as the drug response. The sensor 50 actually measures the urine volume per unit time as the drug response. In other words, the sensor 50 is a urine volume sensor.

[0062] The operation of the biological function estimation device 20 according to this embodiment will be described with reference to FIG.

[0063] In step S101, the control unit 21 acquires test results before drug administration. Specifically, the control unit 21 retrieves data for renal function evaluation, such as blood test values ​​and electronic medical record data included in the in-hospital data, from the database 30 via the communication unit 23. As a modified example, the control unit 21 may accept input of data for renal function evaluation from a user, such as a doctor or nurse, via the input unit 24, such as a touch screen. Alternatively, the control unit 21 may accept input of the presence or absence of a disease to be considered in renal function evaluation, such as thyroid dysfunction or the presence of some malignant tumors, via the electronic medical record data or the input unit 24, such as a touch screen.

[0064] In step S102, the control unit 21 evaluates the renal function of the patient before drug administration from the test results acquired in step S101. That is, the control unit 21 calculates an evaluation value Ve from the test results before drug administration. For example, the control unit 21 calculates GFR or eGFR by substituting the blood test values ​​and the electronic medical record data values ​​included in the data acquired in step S101 into a formula for evaluating renal function. Alternatively, the control unit 21 calculates GFR by multiplying the eGFR included in the data acquired in step S101 by a coefficient. In the former calculation, for information that is not quantified, such as the presence or absence of a disease, information in a normalized or quantified state that can be substituted into the formula is substituted into the formula. The control unit 21 stores the obtained evaluation value Ve in the storage unit 22.

[0065] After step S102, if the number of drug administrations has not reached the threshold value Vt, the process of step S103 is performed. The threshold value Vt may be any value equal to or greater than 1, but is set to a value equal to or greater than 2 in this embodiment.

[0066] In step S103, the control unit 21 acquires data Dm of a drug to be administered to the patient. Specifically, the control unit 21 acquires data Dm of a drug to be administered from the pump 40 or a device that monitors and controls the pump 40, via the communication unit 23. As a modified example, the control unit 21 may accept input of the drug data Dm from a user, such as a doctor or a nurse, via the input unit 24, such as a touch screen.

[0067] In step S104, the control unit 21 predicts the urine volume per unit time as a drug reaction related to the renal function of the patient based on the evaluation value Ve obtained in step S102 and the drug data Dm acquired in step S103. Specifically, the control unit 21 refers to a table pre-stored in the storage unit 22 to identify a predicted value Vp corresponding to a combination of the evaluation value Ve and at least one of the drug type, total administration amount, and dosage per unit time included in the drug data Dm. Alternatively, the control unit 21 calculates the predicted value Vp by substituting the evaluation value Ve and at least one of the drug type, total administration amount, and dosage per unit time included in the drug data Dm into a pre-defined formula.

[0068] In step S105, the control unit 21 acquires the result of actual measurement of the urine volume per unit time by the sensor 50 as a drug reaction related to the renal function of the patient after drug administration for a certain period of time by the pump 40. Specifically, the control unit 21 acquires the actual measurement value Vm from the sensor 50 via the communication unit 23. As a modified example, the control unit 21 may accept input of the actual measurement value Vm from a user such as a doctor or nurse via the input unit 24 such as a touch screen.

[0069] In step S106, the control unit 21 corrects the evaluation value Ve obtained in step S102 according to the comparison result between the predicted value Vp obtained in step S104 and the actual measurement value Vm obtained in step S105. Specifically, the control unit 21 changes at least one of the formula and the coefficient for calculating the evaluation value Ve in step S102 according to the comparison result between the predicted value Vp and the actual measurement value Vm. For example, the control unit 21 corrects the formula for renal function evaluation applied in step S102 so that the difference between the predicted value Vp and the actual measurement value Vm becomes small. Alternatively, when there are multiple formulas for renal function evaluation applicable in step S102, the control unit 21 reselects the formula to be used so that the difference between the predicted value Vp and the actual measurement value Vm becomes small. Alternatively, the control unit 21 corrects the coefficient by which the eGFR is multiplied in step S102 so that the difference between the predicted value Vp and the actual measurement value Vm becomes small. Then, in step S102, the control unit 21 recalculates the evaluation value Ve using at least one of the corrected formula and the coefficient. For example, the control unit 21 substitutes the blood test values ​​and the electronic medical record data values ​​included in the data acquired in step S101 into the corrected formula for evaluating renal function to recalculate the GFR or eGFR. That is, the control unit 21 corrects the GFR or eGFR. Alternatively, the control unit 21 multiplies the eGFR included in the data acquired in step S101 by the corrected coefficient to recalculate the GFR. That is, the control unit 21 corrects the GFR. The control unit 21 overwrites the evaluation value Ve stored in the memory unit 22 with the corrected value.

[0070] After step S102, if the number of drug administrations reaches the threshold value Vt, the process of step S107 is performed.

[0071] In step S107, the output unit 25 outputs the corrected value of the evaluation value Ve. Specifically, the control unit 21 causes the output unit 25, such as a display, to display the corrected value of the evaluation value Ve that has been overwritten and saved in the storage unit 22.

[0072] In this embodiment, since the threshold value Vt is set to a value of 2 or more, the drug is administered to the patient at multiple time points. The control unit 21 predicts the drug response for each of the drug administrations at the multiple time points, and corrects the evaluation value Ve according to a comparison result between the obtained new predicted value Vp and a new measured value Vm obtained by actually measuring the drug response after the drug administration. Specifically, the control unit 21 predicts the drug response for the drug administration at a second time point after the first time point among the multiple time points, based on the corrected value of the evaluation value Ve obtained for the drug administration at the first time point and the data Dm of the drug administered to the patient at the second time point, and further corrects the evaluation value Ve according to a comparison result between the obtained new predicted value Vp and a new measured value Vm obtained by actually measuring the drug response after the drug administration.

[0073] For example, the control unit 21 performs the process from step S103 to step S106 for the drug administration at time T1 to correct the evaluation value Ve. As a result, in step S102, the corrected value of the evaluation value Ve is obtained. The control unit 21 performs the process from step S103 to step S106 for the drug administration at the next time T2 to correct the corrected value of the evaluation value Ve. That is, the control unit 21 further corrects the evaluation value Ve. As a result, in step S102, the further corrected value of the evaluation value Ve is obtained. When the threshold value Vt=n and n>2, the control unit 21 performs the process from step S103 to step S106 for the drug administration at each of the next time points T3 to Tn to further correct the evaluation value Ve. In this way, in this embodiment, the control unit 21 recursively corrects the evaluation value Ve.

[0074] As one variant of this embodiment, the control unit 21 may record the corrected values ​​of the evaluation value Ve obtained for drug administration at each of multiple time points as provisional correction values, and further correct the evaluation value Ve using the recorded multiple provisional correction values.

[0075] For example, the control unit 21 performs the processes from step S103 to step S106 for drug administration at one or more time points to recursively correct the evaluation value Ve. In step S107, the control unit 21 records the value of the evaluation value Ve after the recursive correction as a provisional correction value C1. The control unit 21 also performs the processes from step S103 to step S106 for drug administration at another one or more time points to recursively correct the evaluation value Ve. In step S107, the control unit 21 records the value of the evaluation value Ve after the recursive correction as a provisional correction value C2. As a result, a plurality of provisional correction values ​​are recorded. The number of provisional correction values ​​may be three or more. The control unit 21 uses these multiple provisional correction values ​​to determine the final output value of the evaluation value Ve. That is, the control unit 21 further corrects the evaluation value Ve. As a method for determining the final output value, any method such as average value calculation, median value calculation, simple regression analysis, multiple regression analysis, polynomial regression, or regularization can be used.

[0076] As described above, in this embodiment, the control unit 21 of the biological function estimation device 20 predicts the patient's urine volume based on the evaluation value Ve obtained by evaluating the patient's renal function from the test results before drug administration and the data Dm of the drug to be administered to the patient. The control unit 21 corrects the evaluation value Ve according to the result of comparison between the obtained predicted value Vp and the actual measurement value Vm obtained by actually measuring the urine volume after drug administration.

[0077] According to this embodiment, the renal function of a patient can be evaluated with high accuracy without the need for testing each time.

[0078] In this embodiment, whether to perform the process of step S103 or the process of step S107 is selected depending on whether the number of drug injections has reached a threshold value Vt, but instead of the number of drug injections, the difference between the predicted value Vp and the actual value Vm may be used. That is, if the difference between the predicted value Vp and the actual value Vm is greater than a specific threshold, the process of step S103 may be selected, and if it is equal to or less than the specific threshold, the process of step S107 may be selected.

[0079] Alternatively, both the threshold value Vt for the number of drug administrations and the threshold value for the difference between the predicted value Vp and the actual measurement value Vm may be used. In this case, the condition for selecting the process of step S107 may be set as the case where both threshold values ​​are satisfied, or the process of step S107 may be selected if the threshold value Vt for the number of drug administrations is satisfied when the difference between the predicted value Vp and the actual measurement value Vm is not smaller than a specific threshold. In addition, even if the threshold value Vt for the number of drug administrations is not satisfied, the process of step S107 may be selected when the difference between the predicted value Vp and the actual measurement value Vm becomes smaller than a specific threshold. By combining the two threshold values ​​in this way, it is possible to obtain a highly accurate renal function evaluation value while setting the upper limit of the time required for the renal function evaluation value to be output according to the usage scene.

[0080] The most accurate index of renal function evaluation is GFR, which can be obtained by measuring the rate of inulin excretion from the glomerulus. However, inulin does not exist in the body. Therefore, inulin must be administered intravenously to measure it, and strict urine collection or complete urination and frequent blood sampling are required. Thus, measuring GFR is very complicated and involves a burden on the patient. Therefore, GFR is not used in clinical practice except in situations where strict evaluation of renal function is required, such as in patients with renal disease or kidney transplant donors.

[0081] Currently, alternative markers are used. eCCr is used, which is calculated by substituting the serum concentration of creatinine, which is excreted from the body, and body weight, age, and sex into an estimation formula. Alternatively, eGFR is used, which is calculated from serum Cre, age, and sex. "eCCr" is estimated creatinine clearance. There is actual CCr as an actual measurement value, but measuring CCr requires blood sampling and accurate measurement of 24-hour urine volume. Therefore, CCr is rarely measured in clinical practice to obtain medication guidelines. "CCr" is creatinine clearance.

[0082] As a result, renal function evaluation in clinical practice involves measuring serum Cre and calculating eCCr or eGFR. However, since these require blood tests, there is a limit to how often they can be performed, and there is a problem that real-time renal function evaluation cannot be performed. Furthermore, there is a problem that the calculated eCCr or eGFR is not accurate. Unlike inulin, creatinine is secreted not only by glomerular filtration but also by the renal tubules, so CCr is about 20% to 30% higher than GFR. In addition, for serum Cre, in sarcopenic patients with reduced muscle mass and in many elderly people, the decline in renal function is not detected. In other words, it is expressed as having good renal function. It is said that more than 50% of people over 85 years old have sarcopenia. The number of people with the disease in Japan is approximately 3.7 million. In addition, in the ICU, even young patients will be bedridden for a while, so muscle mass will decrease. If eCCr or eGFR is calculated using serum Cre for such patients, the evaluation value of renal function will be calculated higher than the true value. "ICU" is an abbreviation for intensive care unit.

[0083] If medication is administered without accurately assessing renal function, there is a risk of side effects occurring when administering a normal dose of a renally excreted drug to a patient with reduced renal function, or of developing drug-induced nephropathy. In particular, when the calculated eCCr or eGFR is around 60mL / min to 100mL / min, where it is difficult to tell that the value is excessive, it becomes difficult to design the dosage. For example, appropriate renal function evaluation is particularly necessary for patients with CKD. "CKD" is an abbreviation for chronic kidney disease.

[0084] Other methods for evaluating renal function that are used in clinical practice include Cys-C and BUN / Cre, but like Cre, eCCr, and eGFR, they require blood tests each time and are therefore not suitable for evaluating renal function in real time.In addition, there are issues with the accuracy of the measurements.

[0085] Serum cystatin C is expressed throughout the body, and its production amount and rate are constant and are not easily affected by external factors such as inflammation and age. Moreover, after extracellular secretion, 99% of cystatin C is reabsorbed from the proximal tubule and catabolized within a short time, and does not recirculate in the blood. Therefore, serum Cys-C is considered to be determined by the glomerular filtration rate. However, it is known that the prediction accuracy of eGFR decreases in patients with end-stage renal failure. It has also been suggested that serum Cys-C is higher than the true value in patients receiving high-dose or long-term steroid administration. Furthermore, serum Cys-C is said to be affected by thyroid dysfunction and the presence of some malignant tumors.

[0086] As for BUN / Cre, since serum CCr is used, it has the same issues as serum CCr.

[0087] According to the present embodiment, by employing one or more feedback systems, renal function can be more accurately estimated without the use of blood tests, and since the feedback is provided in real time, renal function of hospitalized patients can be accurately and in real time.

[0088] Second embodiment The differences from the first embodiment will be mainly described.

[0089] In this embodiment, as shown in Fig. 5, in order to improve the accuracy of the evaluation of renal function, an optimal drug administration pattern for correcting the evaluation value Ve is estimated, and drug administration is instructed according to the estimated pattern. As the drug administration pattern, at least one of a pattern in which the drug is gradually increased from a low dose and a pattern in which the drug is gradually decreased from a high dose within the range of the determined prescription is selected. Therefore, the total amount of the drug administered is as prescribed.

[0090] The operation of the biological function estimation device 20 according to this embodiment will be described with reference to FIG.

[0091] The processing from step S201 to step S206 is the same as the processing from step S101 to step S106 in the first embodiment, and therefore the description thereof will be omitted.

[0092] In step S207, the control unit 21 adjusts the pattern of drug administration from the next time onwards, depending on the comparison result between the predicted value Vp obtained in step S204 and the actual value Vm obtained in step S205. Specifically, the control unit 21 controls the pump 40 to set at least one of the number of drug administrations and the dosage per unit time, thereby setting the drug administration pattern to the above-mentioned pattern.

[0093] In this embodiment, the control unit 21 adjusts the drug administration pattern in step S207 by referring to a plurality of drug administration patterns pre-stored in the storage unit 22. Even if the predicted value Vp obtained in step S202 continues to fluctuate and the difference between the predicted value Vp and the actual measured value Vm does not converge even if the processes from step S202 to step S206 are repeated, the optimal drug administration pattern for obtaining a stable predicted value Vp is selected.

[0094] When the predicted value Vp obtained in step S202 varies greatly each time the processing from step S202 to step S206 is performed, the main cause of this may be that the magnitude of the actual measurement value Vm obtained in step S205 is not appropriate. If the urine volume per unit time as the actual measurement value Vm is too small, the measurement value is likely to vary due to measurement errors, and the predicted value Vp calculated based on this varies. On the other hand, if the urine volume per unit time is too large, the fluctuation in the urine volume occurring while the processing from step S201 to step S206 is repeated also becomes large, so the predicted value Vp varies greatly. The urine volume per unit time is determined by the balance between renal function and drug dosage. Therefore, by administering a drug in a stepwise increase from a low dose or a stepwise decrease from a high dose within the range of the prescribed prescription, a measurement point where the urine volume is not too small and not too large can be obtained, and as a result, a stable predicted value Vp can be obtained.

[0095] A plurality of drug administration patterns are stored in the memory unit 22, and the drug can be increased stepwise from a low dose, or decreased stepwise from a high dose. For example, even when the drug is increased stepwise from a low dose, various drug administration patterns having different lengths of time for maintaining the low dose and the high dose can be selected. By using the optimal drug administration pattern, it is possible to obtain more measurement points at which the actual measured value Vm that is optimal for obtaining the predicted value Vp can be obtained within a range that does not suppress the required drug effect.

[0096] As the drug administration pattern in step S207, an optimal one is automatically selected based on the difference between the predicted value Vp obtained in step S204 and the actual value Vm obtained in step S205, the magnitude of the actual value Vm itself, and the amount of fluctuation of the actual value Vm. In addition, based on the patient condition or past performance, a drug administration pattern to be used or a drug administration pattern not to be used can be specified via input unit 24 such as a touch screen, or a drug administration pattern to be used preferentially can be set in advance.

[0097] In this embodiment, the control unit 21 stores the actual measurement value Vm in the storage unit 22 and calculates the amount of fluctuation in the actual measurement value Vm. As a method for calculating the amount of fluctuation, a commonly used calculation method such as a method of determining the difference from the previous value, a method of using the time derivative of the actual measurement value Vm, or a method of evaluating the standard deviation or variation can be used.

[0098] The process of step S208 is the same as the process of step S107 in the first embodiment, and therefore a description thereof will be omitted.

[0099] As described above, in this embodiment, the control unit 21 sets a pattern of drug administration at each of the multiple time points. Specifically, the control unit 21 adjusts the pattern of drug administration at the next time point and thereafter, depending on a comparison result between a new predicted value Vp and a new actual value Vm obtained for drug administration at any one of the multiple time points.

[0100] According to this embodiment, even if the actual measured value Vm is temporarily not within the appropriate range for obtaining the predicted value Vp and the predicted value Vp continues to fluctuate, an optimal drug administration pattern is selected to obtain a stable predicted value Vp, thereby improving the estimation accuracy and measurement efficiency of renal function.

[0101] In this embodiment, whether to perform the process of step S203 or the process of step S208 is selected depending on whether the number of drug injections has reached the threshold value Vt, but as in the first embodiment, the difference between the predicted value Vp and the actual value Vm may be used instead of the number of drug injections. Alternatively, both the threshold value Vt for the number of drug injections and the threshold value for the difference between the predicted value Vp and the actual value Vm may be used. In either case, an appropriate drug injection pattern is selected by the process of step S207 based on the difference between the predicted value Vp and the actual value Vm, or the magnitude or amount of fluctuation of the actual value Vm itself.

[0102] As a modification of this embodiment, the control unit 21 may select, from a group of predetermined pattern candidates, a different pattern candidate for each time point as the drug administration pattern.

[0103] As a modification of this embodiment, when setting a drug administration pattern, the control unit 21 may refer to performance data Da indicating a history of past drug administrations and at least one of the predicted value Vp, the actual value Vm, and the corrected value of the evaluation value Ve. The history indicated in the performance data Da may include the history of other patients.

[0104] Third embodiment The differences from the first embodiment will be mainly described.

[0105] In this embodiment, as shown in Fig. 7, in order to improve the accuracy of the evaluation of renal function, an optimal drug administration pattern for correcting the evaluation value Ve is estimated, drug administration is instructed according to the estimated pattern, and the timing of urine volume measurement is instructed. The drug administration pattern includes a procedure for increasing or decreasing the dosage within a prescribed range, as well as a bolus administration in which a fixed amount is rapidly administered, or a procedure for stopping administration, and a pattern is selected in which the procedure for performing measurement by the sensor 50 in accordance with these changes in the drug administration pattern is repeated.

[0106] The operation of the biological function estimation device 20 according to this embodiment will be described with reference to FIG.

[0107] In this embodiment, a plurality of combinations of an estimation formula for calculating an evaluation value Ve from a comparison result between the predicted value Vp and the actual value Vm in step S302 and a drug administration pattern corresponding to the estimation formula are stored in the storage unit 22. Then, by selecting an optimal drug administration pattern, a more accurate evaluation value Ve can be obtained.

[0108] The processing from step S301 to step S306 is the same as the processing from step S101 to step S106 in the first embodiment, and therefore the description thereof will be omitted.

[0109] In step S307, the control unit 21 adjusts the pattern of drug administration from the next time onwards, depending on the comparison result between the predicted value Vp obtained in step S304 and the actual value Vm obtained in step S305. Specifically, the control unit 21 controls the pump 40 to set at least one of the number of drug administrations and the dosage per unit time, and sets the drug administration timing to set the drug administration pattern to the above-mentioned pattern. The drug administration timing may be set not only to the time of drug administration, but also to whether or not to stop administration of the drug during continuous administration, and whether or not to re-administer the drug if it is stopped.

[0110] The amount of urine per unit time is determined by the balance between renal function and the amount of drug administered. When the drug is a diuretic, the amount of urine per unit time generally increases by increasing the amount administered, but the tendency of increase varies. The tendency of increase is expressed by the slope or shape of the plot when the diuretic dosage is plotted on the horizontal axis and the amount of urine on the vertical axis, the continuity of change, or the presence or absence of a threshold. The tendency of increase is closely related to the renal function to be sought. Therefore, within the range of a prescribed prescription, a combination of the predicted value Vp and the actual value Vm for the urine volume responsiveness when a diuretic is administered in a bolus, a combination of the predicted value Vp and the actual value Vm for the urine volume responsiveness when the administration is stopped, and a combination of the predicted value Vp and the actual value Vm for the fluctuation in the urine volume when the dosage is intentionally increased or decreased will each provide a different convergence pattern of the difference between the predicted value Vp and the actual value Vm. Then, three different evaluation values ​​Ve can be obtained from each. Since the degree of convergence of the difference between the predicted value Vp and the actual measurement value Vm differs depending on the drug administration pattern, the estimation accuracy as a renal function evaluation value differs among the three evaluation values ​​Ve. Therefore, by preparing a formula for calculating the evaluation value Ve from the predicted value Vp and the actual measurement value Vm and a plurality of combinations of the drug administration patterns corresponding thereto, it is possible to select the drug administration pattern that sufficiently reduces the difference between the predicted value Vp and the actual measurement value Vm when the processes from step S302 to step S308 are repeated as the optimal drug administration pattern for obtaining a highly accurate evaluation value Ve, and to obtain the desired renal function evaluation value.

[0111] In addition, by preparing a formula for calculating the evaluation value Ve from the predicted value Vp and the actual measured value Vm, and a plurality of combinations of the corresponding drug administration patterns, when calculating the evaluation value Ve using a machine learning method, the plurality of combinations can be used as different explanatory variables. The explanatory variables are input information for estimating the evaluation value Ve. The plurality of combinations can be treated as individual regression trees in the random forest method, or can be used as penalty terms in machine learning. These are particularly effective methods when the difference between the predicted value Vp and the actual measured value Vm is sufficiently small and converges to obtain the evaluation value Ve for each of the plurality of drug administration patterns, but the obtained evaluation values ​​Ve are different, and it is not possible to determine which evaluation value Ve should be used.

[0112] When the processes from step S302 to step S308 are repeated, combinations of predicted values ​​Vp and actual values ​​Vm are measured for various drug administration patterns. The obtained combinations of predicted values ​​Vp, actual values ​​Vm, and evaluation values ​​Ve are stored in the storage unit 22 in association with the drug administration patterns. In this case, the combinations of predicted values ​​Vp, actual values ​​Vm, and evaluation values ​​Ve may be stored in the storage unit 22 as time-series data in association with the drug administration patterns.

[0113] In step S308, the control unit 21 adjusts the timing of measuring the urine volume as a drug response according to the drug administration pattern adjusted in step S307. Specifically, the control unit 21 controls the sensor 50 to adjust the timing of measuring the urine volume to match the drug administration pattern adjusted in step S307. In this embodiment, when the drug administration pattern is changed, accurate measurement of the urine volume responsiveness corresponding to the administration pattern affects the estimation accuracy of the evaluation value Ve. Therefore, by performing the urine volume measurement so that the timing of drug administration and the urine volume measurement are aligned, it is possible to obtain a highly accurate evaluation value Ve.

[0114] The process in step S309 is the same as the process in step S107 in the first embodiment, and therefore a description thereof will be omitted.

[0115] According to this embodiment, it is possible to estimate renal function using the urinary volume responsiveness when drug administration patterns are changed in various ways, thereby improving the accuracy of estimating renal function.

[0116] In this embodiment, whether to perform the process of step S303 or the process of step S309 is selected depending on whether the number of drug administrations has reached the threshold value Vt. However, as in the first embodiment, instead of the number of drug administrations, the difference between the predicted value Vp and the actual value Vm may be used. Alternatively, both the threshold value Vt for the number of drug administrations and the threshold value for the difference between the predicted value Vp and the actual value Vm may be used. In either case, measurements may be performed by changing multiple types of drug administration patterns according to a predetermined pattern, or a drug administration pattern may be automatically selected from several candidates stored in the storage unit 22 in advance based on the value or variation of the evaluation value Ve obtained after drug administration. In addition, the same drug administration pattern may be repeated to reduce the variation in the combination of the predicted value Vp and the actual value Vm obtained for a specific drug administration pattern and the evaluation value Ve obtained thereafter.

[0117] Furthermore, as a result of performing the processes from step S302 to S308 for several drug administration patterns, in the case where the predicted value Vp converges for each drug administration pattern, but the difference between the predicted value Vp and the actual value Vm is not sufficiently small, the final evaluation value Ve may be obtained using the multiple evaluation values ​​Ve obtained for each drug administration pattern. In this case, any method may be selected as a method for obtaining the final evaluation value Ve. For example, the average or median of the multiple evaluation values ​​Ve may be calculated as the final evaluation value Ve, or the final evaluation value Ve may be obtained by simple regression analysis, multiple regression analysis, polynomial regression, or regularization.

[0118] The storage unit 22 can also record in association with the test results used in the past renal function evaluation of different patients, in-hospital data such as the presence or absence of underlying diseases, and drug administration patterns that were effective in reducing the difference between the predicted value Vp and the actual value Vm. When selecting the optimal drug administration pattern for the current subject, it is also possible to refer to similar patient information from the test results used in the past renal function evaluation of the current subject and in-hospital data such as the presence or absence of underlying diseases, and to preferentially select a drug administration pattern that was effective in a similar patient in the past. Furthermore, it is also possible to preferentially select a drug administration pattern that was effective in a similar case based on the past data stored in the storage unit 22, using the degree of convergence of the predicted value Vp for each drug administration pattern, or the decreasing tendency of the difference between the predicted value Vp and the actual value Vm itself as reference data.

[0119] (Fourth embodiment) The differences from the third embodiment will be mainly described.

[0120] In this embodiment, as shown in Fig. 9, progress data Dp of drug reaction to administration of a specific drug is recorded. The progress data Dp includes data recording daily changes during the patient's hospitalization and changes from the previous discharge at the time of re-admission. The progress data Dp is referred to when predicting drug reaction related to renal function. The history shown in the progress data Dp may include the history of other patients.

[0121] The operation of the biological function estimation device 20 according to this embodiment will be described with reference to FIG.

[0122] The processing from step S401 to step S403 is the same as the processing from step S301 to step S303 in the third embodiment, and therefore the description thereof will be omitted.

[0123] In step S404, when predicting the urine volume as a drug response, the control unit 21 refers to the progress data Dp indicating the history of past drug administrations and drug responses. Specifically, the control unit 21 analyzes the history indicated by the progress data Dp to extract the characteristics of the drug response specific to the patient. When the control unit 21 specifies the predicted value Vp by referring to a table pre-stored in the storage unit 22, the control unit 21 corrects the specified predicted value Vp to match the characteristics of the drug response specific to the patient. Alternatively, when the control unit 21 calculates the predicted value Vp by applying a pre-defined formula, the control unit 21 corrects the calculated predicted value Vp to match the characteristics of the drug response specific to the patient.

[0124] The process of step S405 is the same as the process of step S305 in the third embodiment, and therefore a description thereof will be omitted.

[0125] In step S406, the control unit 21 records the date and time of drug administration by the pump 40, the type and amount of the administered drug, and the actual measurement value Vm obtained in step S405 as part of the progress data Dp.

[0126] The processing from step S407 to step S410 is the same as the processing from step S306 to step S309 in the third embodiment, and therefore a description thereof will be omitted.

[0127] According to this embodiment, when obtaining the predicted value Vp, various data for renal function evaluation including the actual measurement value Vm, drug data Dm, and progress data Dp can be referred to, so that a predicted value Vp closer to the true value can be obtained. Therefore, the number of times the processes from step S402 to step S409 are repeated can be reduced, and the time required for outputting the renal function evaluation value in step S410 can be shortened.

[0128] Reducing the number of times the processes from step S402 to step S409 are repeated and shortening the time required to output the renal function evaluation value in step S410 is important in situations where it is necessary to evaluate the renal function of a patient in real time, such as in an ICU.

[0129] In other words, according to this embodiment, the prediction accuracy of the predicted value Vp based on the drug response is improved, so that the time required to output the renal function evaluation value can be shortened, and the estimation accuracy of renal function is maintained or improved even in usage methods such as real-time monitoring.

[0130] As a modified example of this embodiment, the control unit 21 may update the progress data Dp every time the predicted value Vp and the actual measured value Vm are obtained, thereby making it possible to obtain a predicted value Vp that is closer to the true value.

[0131] If the control unit 21 stores combinations of predicted values ​​Vp and actual values ​​Vm for various administration patterns applied in the past as progress data Dp indicating the history of past drug administrations and drug reactions, it can perform machine learning using these to obtain a predicted value Vp that is more individually optimized for the measurement subject.

[0132] In this embodiment, whether to perform the process of step S403 or the process of step S410 is selected depending on whether the number of drug administrations has reached the threshold value Vt, but as in the first embodiment, the difference between the predicted value Vp and the actual value Vm may be used instead of the number of drug administrations. Alternatively, both the threshold value Vt for the number of drug administrations and the threshold value for the difference between the predicted value Vp and the actual value Vm may be used. In either case, when the processes of steps S402 to S409 are repeated, the progress data Dp is referenced to predict the drug reaction related to renal function.

[0133] 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]

[0134] 10. System 20. Biological function estimation device 21 Control section 22 Memory section 23 Communications Department 24 Input section 25 Output section 30 Databases 40 Pump 50 Sensors

Claims

1. A biological function estimation device having a control unit that predicts a drug response of a patient based on an evaluation value obtained by evaluating the patient's biological function from test results before drug administration and data on the drug to be administered to the patient, and corrects the evaluation value depending on the result of a comparison between the obtained predicted value and an actual measured value obtained by actually measuring the drug response after drug administration.

2. The biological function estimation device according to claim 1 , wherein the control unit changes at least one of a formula and a coefficient for calculating the evaluation value in accordance with the result of the comparison.

3. 3. The biological function estimation device according to claim 1, wherein the data includes at least one of a type of drug, a total administered amount, and an administered amount per unit time.

4. The control unit predicts the drug response for drug administration at each of multiple time points, and corrects the evaluation value based on a comparison result between the obtained new predicted value and a new actual measured value obtained by actually measuring the drug response after drug administration.

5. 5. The biological function estimation device of claim 4, wherein the control unit predicts the drug response for drug administration at a second time point after the first time point among the multiple time points based on a corrected value of the evaluation value obtained for drug administration at the first time point and data on the drug to be administered to the patient at the second time point, and further corrects the evaluation value depending on a comparison result between the obtained new predicted value and a new actual value obtained by actually measuring the drug response after drug administration.

6. The biological function estimation device of claim 4, wherein the control unit records the corrected values ​​of the evaluation value obtained for drug administration at each of the multiple time points as provisional correction values, and further corrects the evaluation value using the recorded multiple provisional correction values.

7. The biological function estimation device according to claim 4 , wherein the control unit sets a pattern of drug administration at each of the plurality of time points.

8. The biological function estimation device of claim 7, wherein the control unit adjusts the pattern at the next time point and thereafter depending on a comparison result between a new predicted value obtained for drug administration at any one of the multiple time points and a new actual measured value.

9. The biological function estimation device according to claim 7 , wherein the control unit selects, as the pattern, a different pattern candidate for each time point from a predetermined group of pattern candidates.

10. 10. The biological function estimation device according to claim 7, wherein the control unit sets the pattern by setting at least one of the number of times a drug is administered and the dosage per unit time.

11. The biological function estimation device according to claim 7 , wherein the control unit sets the pattern by setting a timing of administration of a drug.

12. The biological function estimation device according to any one of claims 7 to 11, wherein the control unit, when setting the pattern, refers to performance data indicating a history of past drug administrations and at least one of predicted values, actual measured values, and corrected values ​​of evaluation values.

13. The biological function estimating apparatus according to claim 12 , wherein the history indicated in the performance data includes histories of other patients.

14. The biological function estimation device according to claim 7 , wherein the control unit adjusts a timing of measuring the drug response in accordance with the pattern.

15. The biological function estimation device according to claim 1 , wherein the control unit refers to progress data indicating a history of past drug administrations and drug reactions when predicting the drug response.

16. The biological function estimating apparatus according to claim 15 , wherein the history indicated in the progress data includes histories of other patients.

17. The biological function estimation apparatus according to claim 1 , wherein the control unit controls a pump that administers a drug to the patient.

18. The biological function estimation device according to claim 1 , wherein the control unit controls a sensor that measures the drug response.

19. The biological function estimation device according to claim 1 , further comprising an output unit that outputs a corrected value of the evaluation value.

20. The biological function estimation device according to claim 1 , wherein the control unit acquires the test results and calculates the evaluation value from the test results.

21. The evaluation value is a value obtained by evaluating renal function as the biological function, The biological function estimation device according to claim 1 , wherein the control unit predicts a urine volume as the drug response.

Images