Control system and control method for target control infusion

Abstract

The invention discloses a control system and method for target control infusion, and relates to the technical field of infusion instruments, and the system comprises a data collection unit which collects personal data; the data processing device is used for predicting a first infusion rate through a data prediction model and simulating the drug concentration change through a pharmacokinetic model to obtain a second infusion rate; the data processing device determines a matching rate according to the first infusion rate and the second infusion rate, and determines the infusion rate according to the confidence rate and the matching rate; the infusion pump is used for infusing medicine to the patient according to the infusion rate. According to the method, the data prediction model and the pharmacokinetic model are respectively utilized to calculate two infusion rates, the two infusion rates respectively reflect the representation and pathological characteristics of the personal data of the patient, the two infusion rates are combined, information in different aspects can be considered, then subjective influence of personnel is abandoned, and the accuracy of the patient is improved. And the accuracy of the infusion rate is greatly improved.

Description

Technical Field

[0001] The present application relates to the technical field of infusion devices, and particularly relates to a control system and a control method for target-controlled infusion. Background Art

[0002] Targeted infusion is a relatively novel medical treatment method. By analyzing the patient's condition and controlling the target-controlled infusion rate, the fluctuation of the drug concentration in the patient's body can be minimized as much as possible, thereby achieving the best treatment effect.

[0003] However, the conditions of different patients are not the same, and even the conditions of the same patient at different times are not the same. In order to achieve the effect of target-controlled infusion, it is necessary to accurately evaluate the patient's condition, and then obtain the personalized and optimal infusion rate. At present, the determination of this target-controlled infusion rate mainly relies on the experience of personnel. Even after a series of methods are used to correct the infusion rate after it is manually determined, since the original infusion rate is affected by humans, the result obtained after correction may also deviate from the optimal value. Summary of the Invention

[0004] Embodiments of the present application provide a control system and a control method for target-controlled infusion, which are used to solve the problem of inaccurate results in accurately determining the infusion rate based on personnel in the prior art.

[0005] On the one hand, embodiments of the present application provide a control system for target-controlled infusion, including:

[0006] A data acquisition unit, configured to acquire personal data of a patient;

[0007] A data processing device, which is built-in with a data prediction model and a pharmacokinetic model. The data prediction model is used to predict a first infusion rate that conforms to the patient's condition according to the personal data, and the pharmacokinetic model is used to simulate the process of the change of the drug concentration in the patient's body according to the personal data, so as to obtain a second infusion rate. The first infusion rate has a corresponding confidence rate;

[0008] The data processing device determines a corresponding matching rate according to the values of the first infusion rate and the second infusion rate, and determines the final infusion rate according to the confidence rate and the matching rate;

[0009] An infusion pump, configured to infuse a drug to the patient according to the infusion rate.

[0010] On the other hand, embodiments of the present application also provide a control method for target-controlled infusion, including:

[0011] Acquiring personal data of a patient;

[0012] A data prediction model is used to predict a first infusion rate that conforms to the patient's condition based on personal data, and the first infusion rate has a corresponding confidence rate;

[0013] A pharmacokinetic model is used to simulate the process of drug concentration change in the patient's body based on personal data to obtain a second infusion rate;

[0014] Determine the corresponding matching rate according to the values of the first infusion rate and the second infusion rate;

[0015] Determine the final infusion rate according to the confidence rate and the matching rate;

[0016] Control the infusion pump according to the infusion rate.

[0017] A control system and a control method for target-controlled infusion in the present application have the following advantages:

[0018] Calculate two infusion rates by using a data prediction model and a pharmacokinetic model respectively. These two infusion rates respectively reflect the appearance and pathological characteristics of the patient's personal data. Combining these two infusion rates can consider information from different aspects, thereby eliminating the subjective influence of personnel and greatly improving the accuracy of the infusion rate. Description of the Drawings

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is a schematic diagram of the composition of a control system for target-controlled infusion provided by an embodiment of the present application. Detailed Embodiments

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0022] Figure 1 It is a schematic diagram of the composition of a control system for target-controlled infusion provided by an embodiment of the present application. An embodiment of the present application provides a control system for target-controlled infusion, including:

[0023] A data acquisition unit for acquiring the personal data of the patient;

[0024] A data processing device is built-in with a data prediction model and a pharmacokinetic model. The data prediction model is used to predict a first infusion rate that suits the patient's condition based on personal data. The pharmacokinetic model is used to simulate the process of the change in the drug concentration in the patient's body based on personal data, and then obtain a second infusion rate. The first infusion rate has a corresponding confidence rate.

[0025] The data processing device determines a corresponding matching rate according to the values of the first infusion rate and the second infusion rate, and determines the final infusion rate according to the confidence rate and the matching rate.

[0026] An infusion pump is used to infuse a drug to a patient according to the infusion rate.

[0027] Exemplarily, the patient's personal data may include age, weight, height, and serum creatinine value. Among them, age, weight, and height can be automatically obtained by the data acquisition unit by accessing the hospital's diagnosis and treatment system, while the serum creatinine value can be obtained by performing a biochemical test on the blood, and this data can also be automatically obtained by the data acquisition unit by accessing the hospital's diagnosis and treatment system.

[0028] The data processing device is built-in with a storage unit. Both the data prediction module and the pharmacokinetic model are stored in the storage unit, and the personal data of the patient obtained by the data acquisition unit will also be temporarily stored in the storage unit.

[0029] In an embodiment of the present application, the data prediction model is established based on a neural network, and the data prediction model is trained using data samples obtained from other patients. The neural network may specifically adopt a convolutional neural network. During the training process, the data samples of other patients include personal data samples and infusion rate samples. When training the data prediction model, the personal data samples are input into the data prediction model to obtain an initial infusion rate, and the parameters of the data prediction model are adjusted according to the difference between the initial infusion rate and the corresponding infusion rate samples to obtain the trained data prediction model.

[0030] Specifically, by training the data prediction model using the data samples of other patients, the data prediction model can learn the relationship between the personal data samples and the infusion rate samples, and then in actual applications, only by inputting the patient's personal data, a relatively accurate first infusion rate can be obtained.

[0031] It should be understood that since there are different relationships between personal data and infusion rates for different drugs, corresponding data prediction models need to be trained for each drug. And during use, the data prediction model of the corresponding drug also needs to be selected.

[0032] During the training process of the data prediction model, the infusion rate can be divided into multiple intervals, and the central value of each interval is used as the classification label for that interval. If an infusion rate sample is within a certain interval, the classification label corresponding to that interval can be used as the label for the infusion rate sample, and then participate in the training process of the data prediction model. Further, the output values of the data prediction model can also be divided into multiple prediction categories, and each prediction category corresponds to a prediction value. After adopting this setting method, the data prediction model can be converted into a multi-classification model. Regardless of the type of personal data input, the predicted category output, that is, the first infusion rate, comes from multiple pre-known prediction values, and these predicted categories all have corresponding confidence rates. The level of the confidence rate represents the credibility of the first infusion rate.

[0033] The first infusion rate and the second infusion rate are obtained by the data prediction module and the pharmacokinetic model respectively. However, since the data prediction model and the pharmacokinetic model are not the same, the data prediction model is trained based on the personal data samples and infusion rate samples of other patients. Therefore, the first infusion rate can only reflect the superficial relationship with the patient's personal data. The pharmacokinetic model starts from the process of drug metabolism in the patient's body to determine the optimal infusion rate. It can be considered that the second infusion rate reflects the pathological relationship with the patient's personal data. When there is a difference between the first infusion rate and the second infusion rate, it means that the infusion rates predicted through the superficial relationship and the pathological relationship of this application are not very accurate. At this time, the matching rate of the first infusion rate and the second infusion rate can be calculated. The matching rate can use the reciprocal of the absolute value of the difference between the first infusion rate and the second infusion rate. If the absolute value of the difference between the two is less than 1, it can be set to 1 and then the reciprocal is calculated. After using the reciprocal calculation, the smaller the difference between the first infusion rate and the second infusion rate, the higher the matching rate.

[0034] In a possible embodiment, when determining the final infusion rate according to the confidence rate and the matching rate, first calculate the sum of the confidence rate of each first infusion rate and the matching rate of each second infusion rate, then determine the largest sum among the multiple sums, and finally use the average value of the first infusion rate and the second infusion rate corresponding to the largest sum as the final infusion rate.

[0035] Exemplarily, the confidence rate reflects the credibility of the first infusion rate. The higher the confidence rate, the higher the accuracy of the first infusion rate. After inputting personal data into the data prediction model in this application, multiple candidate infusion rates are output. Each candidate infusion rate has a corresponding confidence rate. The first infusion rate is selected as the multiple candidate infusion rates with the highest confidence rate. Therefore, the first infusion rate obtained in this application is not one, but many, while the second infusion rate is only one. By calculating the difference between the first infusion rate and the second infusion rate, multiple matching rates can be obtained.

[0036] Based on the fact that the matching rate characterizes the matching degree between the first infusion rate and the second infusion rate, the larger the sum of the matching rate and the confidence rate, the better the corresponding first infusion rate and second infusion rate are both optimal in terms of appearance and pathology. At this time, taking the average value of the corresponding first infusion rate and second infusion rate as the final infusion rate can maximize the accuracy of the infusion rate.

[0037] Furthermore, in addition to taking the average value of the first infusion rate and the second infusion rate as the final infusion rate, the infusion rate can also be determined by weighted summation. When using the weighted summation method, the respective weights of the first infusion rate and the second infusion rate can be set as needed, but it is necessary to ensure that the sum of the two weights is 1. Taking the average value and the weighted summation result of the first infusion rate and the second infusion rate as the infusion rate can simultaneously consider the appearance relationship and the pathological relationship between the patient's personal data and the infusion rate, combining information from different perspectives and improving the accuracy of the infusion rate.

[0038] In a possible embodiment, the method for simulating the second infusion rate through a pharmacokinetic model includes:

[0039] Determine the first-order rate constant according to the following formula:

[0040]

[0041] where, k 1 is the first-order rate constant for the drug to be transported from the central compartment to the peripheral compartment, k 2 is the first-order rate constant for the drug to be transported from the peripheral compartment to the central compartment, k 3 is the first-order rate constant for the drug to be eliminated from the central compartment, Q is the intercompartmental clearance rate, with the unit L / h, V 1 and V 2 are the apparent volume of distribution of the central compartment and the apparent volume of distribution of the peripheral compartment respectively, and the units are both L, and CL is the clearance rate of the central compartment, with the unit L / h.

[0042] Determine the second-order constant according to the following formula:

[0043]

[0044] Among them, a and b are the first secondary constant and the second secondary constant respectively.

[0045]

[0046] Among them, C is the concentration data after drug infusion, with the unit of mg / L, k is the second infusion rate, with the unit of mg / h, e is the natural constant, t is the time used for infusion, with the unit of h.

[0047] Exemplarily, after determining the age, symptoms and drug of the patient, the intercompartmental clearance Q, the apparent volume of distribution in the central compartment V 1 , the apparent volume of distribution in the peripheral compartment V 2 and the central compartment clearance CL can all be obtained by looking up a table. For example, for adults aged between 12 and 65, when meropenem is used, the value of Q is 18.6, and the value of V 1 is V 1 = 10.80×(WT / 70) 0.99 , the value of V 2 is 12.6, and the value of CL is CL = 14.6×(Ccr / 83) 0.62 ×(Age / 35) -0.34 , while for other drugs, such as imipenem, corresponding data need to be taken, where WT is the body weight, Age is the age, and Ccr is the endogenous creatinine clearance rate, with the unit of μmol / L.

[0048] According to the above formula for calculating C, the drug concentration change curve after pharmacokinetic simulation can be obtained. For the same drug, its drug concentration change curve can be fixed, that is, under this concentration change curve, the drug can achieve the best therapeutic effect. Therefore, C can be set as a fixed value, and this fixed value will change with the change of the infusion time, rather than a fixed and unchanging value. When the time is known, the corresponding drug concentration C can be determined based on the curve, and then the second infusion rate k can be calculated.

[0049] For children aged less than or equal to 12 years old, the central compartment clearance CL is determined according to the following formula:

[0050] CL = 4.22×(Ccr / 53.35) 0.29 ×(WT / 13.5) 0.86

[0051] For adults aged greater than 12 years old and less than or equal to 65 years old, the central compartment clearance CL is determined according to the following formula:

[0052] CL = 14.6×(Ccr / 83) 0.62×(Age / 35) -0.34

[0053] For the elderly over 65 years old, the central compartment clearance rate CL is determined according to the following formula:

[0054] CL = 8.98×[1 + 0.0182×(Ccr - 55.3)].

[0055] It should be understood that the above formulas for calculating the central compartment clearance rate CL take meropenem as an example of the drug. If other drugs are used, the corresponding calculation formulas need to be selected.

[0056] In a possible embodiment, the endogenous creatinine clearance rate Ccr is determined according to the following formula:

[0057] Male: Ccr = (140 - Age)×WT / (72×Scr)

[0058] Female: Ccr = (140 - Age)×WT / (85×Scr)

[0059] Wherein, Scr is the serum creatinine value, with the unit of μmol / L.

[0060] Exemplarily, when calculating the endogenous creatinine clearance rate Ccr, the situation of newborns is also considered. Newborns are infants less than 28 days old after birth. For newborns, the formula for calculating the endogenous creatinine clearance rate Ccr is:

[0061] Ccr = K×HT / Scr

[0062] Wherein, K can take 0.55, and HT is the height.

[0063] The embodiment of the present application also provides a control method for target-controlled infusion, and the method includes:

[0064] S100, collecting the personal data of the patient;

[0065] S110, using a data prediction model to predict a first infusion rate that conforms to the patient's situation according to the personal data, and the first infusion rate has a corresponding confidence rate;

[0066] S120, using a pharmacokinetic model to simulate the process of drug concentration change in the patient's body according to the personal data to obtain a second infusion rate;

[0067] S130, determining a corresponding matching rate according to the values of the first infusion rate and the second infusion rate;

[0068] S140, determining the final infusion rate according to the confidence rate and the matching rate;

[0069] S150, control the infusion pump according to the infusion rate.

[0070] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0071] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A control system for target-controlled infusion, characterized in that: include: Data collection unit, used to collect personal data of patients; a data processing device having a built-in data prediction model and a pharmacokinetic model, wherein the data prediction model is used to predict a first infusion rate that meets the patient's condition based on the personal data, and the pharmacokinetic model is used to simulate a process of drug concentration change in the patient's body based on the personal data, thereby obtaining a second infusion rate, wherein the first infusion rate has a corresponding confidence rate; The data processing device determines a corresponding matching rate according to the values ​​of the first infusion rate and the second infusion rate, and determines a final infusion rate according to the confidence rate and the matching rate; An infusion pump is used to infuse drugs into a patient at the infusion rate.

2. A control system for target-controlled infusion according to claim 1, characterized in that: The data prediction model is established based on a neural network and is trained using data samples obtained from other patients.

3. A control system for target-controlled infusion according to claim 2, characterized in that: The data samples of other patients include personal data samples and infusion rate samples. When training the data prediction model, the personal data samples are input into the data prediction model to obtain an initial infusion rate. The parameters of the data prediction model are adjusted according to the difference between the initial infusion rate and the corresponding infusion rate samples to obtain a trained data prediction model.

4. A control system for target-controlled infusion according to claim 1, characterized in that: After the personal data is input into the data prediction model, a plurality of candidate infusion rates are output, each of which has a corresponding confidence rate, and the plurality of candidate infusion rates with the highest confidence rates are selected as the first infusion rate.

5. A control system for target-controlled infusion according to claim 1, characterized in that: The matching ratio is the inverse of the difference between the first infusion rate and the second infusion rate.

6. A control system for target-controlled infusion according to claim 1, characterized in that: When determining the final infusion rate according to the confidence rate and the matching rate, first calculate the sum of the confidence rate of each first infusion rate and the matching rate of the second infusion rate, then determine the largest sum among the multiple sums, and finally take the average value of the first infusion rate and the second infusion rate corresponding to the largest sum as the final infusion rate.

7. A control system for target-controlled infusion according to claim 1, characterized in that: The method for obtaining the second infusion rate by simulating the pharmacokinetic model includes: The first-order rate constant was determined as follows: Wherein, k1 is the first-order rate constant for drug transport from the central compartment to the peripheral compartment, k2 is the first-order rate constant for drug transport from the peripheral compartment to the central compartment, k3 is the first-order rate constant for drug elimination from the central compartment, Q is the inter-compartmental clearance rate, in L / h, V1 and V2 are the apparent distribution volume of the central compartment and the apparent distribution volume of the peripheral compartment, respectively, in L, CL is the central compartment clearance rate, in L / h; Determine the secondary constant as follows: Among them, a and b are the first and second order constants, respectively; Wherein, C is the concentration data after drug infusion, in mg / L, k is the second infusion rate, in mg / h, e is a natural constant, and t is the time taken for infusion, in h.

8. A control system for target-controlled infusion according to claim 7, characterized in that: The central compartment clearance CL is determined based on the endogenous creatinine clearance Ccr and the personal data. The endogenous creatinine clearance Ccr is determined according to the following formula: Male: Ccr=(140-Age)×WT / (72×Scr) Female: Ccr=(140-Age)×WT / (85×Scr) Wherein, Age is age, WT is weight, Scr is blood creatinine value, and the unit is μmol / L. The personal data includes age, weight and blood creatinine value.

9. A method for controlling a target-controlled infusion system according to any one of claims 1 to 8, characterized in that: include: Collecting personal data of patients; Using a data prediction model to predict a first infusion rate that matches the patient's condition based on the personal data, the first infusion rate having a corresponding confidence rate; Using a pharmacokinetic model to simulate the concentration change process of the drug in the patient's body according to the personal data to obtain a second infusion rate; determining a corresponding matching rate according to the values ​​of the first infusion rate and the second infusion rate; determining a final infusion rate according to the confidence rate and the matching rate; The infusion pump is controlled according to the described infusion rate.

Images