Anesthesia target control parameter optimization method and device, electronic equipment and storage medium

Abstract

The invention discloses an anesthesia target control parameter optimization method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: constructing a pharmacokinetic model; determining a deviation value and a deviation change rate between a first anesthesia target control parameter output by the pharmacokinetic model and a target anesthesia target control parameter; determining an initial weight and a threshold value of the fuzzy neural network based on the radial basis function; inputting the deviation value and the deviation change rate into a radial basis function-based fuzzy neural network provided with an initial weight and a threshold value to obtain a control parameter of a PID controller; and controlling the pharmacokinetic model to output a second anesthesia target control parameter by using the PID controller based on the control parameter in combination with the deviation value and the deviation change rate. According to the method, the optimized initial weight and threshold are determined, the accuracy of the initial parameters of the fuzzy neural network is improved, the control parameters are optimized, the adaptive capacity of the PID controller is enhanced, and more accurate anesthesia target control parameters can be obtained.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an anesthesia target control parameter optimization method, device, electronic device and storage medium. Background Art

[0002] With the development of modern medical technology, anesthesia technology plays a vital role in surgery. The purpose of anesthesia is to provide patients with a painless and safe surgical experience while ensuring the stability of their vital signs during surgery. In order to achieve this goal, the Target Controlled Infusion (TCI) technology of anesthesia came into being. TCI technology uses a computer-assisted system to automatically adjust the infusion rate of anesthetic drugs based on the pharmacokinetic and pharmacodynamic models of the drugs to maintain the drug concentration in the patient's body within the preset target range, thereby controlling the depth of anesthesia.

[0003] However, the anesthesia target control parameters of current anesthesia target controlled infusion technology are still not accurate enough. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose an anesthesia target control parameter optimization method, device, electronic device and storage medium to optimize the anesthesia target control parameters and improve the accuracy of the anesthesia target control parameters.

[0005] To achieve the above object, an embodiment of the present application provides an anesthesia target control parameter optimization method, the method comprising the following steps:

[0006] Constructing pharmacokinetic models;

[0007] Outputting a first anesthesia target control parameter using the pharmacokinetic model;

[0008] Determining a deviation value between the first anesthesia target control parameter and a target anesthesia target control parameter and a rate of change of the deviation value over time as a deviation change rate;

[0009] Determine the initial weights and thresholds of the fuzzy neural network based on radial basis function;

[0010] Inputting the deviation value and the deviation change rate into the radial basis function-based fuzzy neural network provided with the initial weight and the threshold value to obtain control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters;

[0011] The PID controller is used to control the pharmacokinetic model to output a second anesthesia target control parameter based on the control parameter and in combination with the deviation value and the deviation change rate.

[0012] In some embodiments, constructing a pharmacokinetic model comprises the following steps:

[0013] The pharmacokinetic model is constructed based on the physiological pharmacodynamic model of propofol in the human body; wherein the physiological pharmacodynamic model of propofol in the human body includes a vein model, a lung model, a brain model, a liver model, a fast equilibrium tissue model, a slow equilibrium tissue model and a deep distribution model.

[0014] In some embodiments, constructing the pharmacokinetic model based on the physiological pharmacodynamics model of propofol in humans comprises the following steps:

[0015] Establishing a third-order differential equation of the lung model, and then determining a transfer function expression of the lung model;

[0016] Establishing a second-order differential equation of the brain model, and then determining a transfer function expression of the brain model;

[0017] Establishing a first-order differential equation of the liver model, and then determining a transfer function expression of the liver model;

[0018] Establishing a first-order differential equation of the fast equilibrium tissue model, and then determining a transfer function expression of the fast equilibrium tissue model;

[0019] Establishing a second-order differential equation of the slow equilibrium tissue model, and then determining a transfer function expression of the slow equilibrium tissue model;

[0020] Establishing a first-order differential equation of the vein model, and then determining a transfer function expression of the vein model;

[0021] Determining a transfer function expression of the deep distribution model;

[0022] The pharmacokinetic model is constructed according to the transfer function expressions corresponding to the venous model, the lung model, the brain model, the liver model, the fast equilibrium tissue model, the slow equilibrium tissue model and the deep distribution model.

[0023] In some embodiments, before determining the initial weights and thresholds of the radial basis function-based fuzzy neural network, the method further comprises the following steps:

[0024] Constructing multiple input layers; wherein each of the input layers is used to input multiple target variables, and the target variables include the deviation value and the deviation change rate;

[0025] Constructing a plurality of fuzzification layers; wherein each of the fuzzification layers is used to divide each of the target variables into fuzzy sets equal in number to the fuzzification layers and calculate the membership degree of each of the target variables to each of the fuzzy sets;

[0026] Constructing multiple fuzzy reasoning layers; wherein each of the fuzzy reasoning layers is used to implement fuzzy reasoning according to the output of each of the fuzzification layers;

[0027] Constructing a plurality of output layers; wherein each of the output layers is used to output a target parameter, and the target parameter includes the control parameter;

[0028] The input layers, the fuzzification layers, the fuzzy inference layers and the output layers are connected in sequence to obtain the fuzzy neural network based on radial basis function.

[0029] In some embodiments, constructing multiple blurring layers comprises the following steps:

[0030] Utilizing each of the fuzzification layers to calculate the membership of each of the target variables to each of the fuzzy sets according to a membership function;

[0031] The membership function is:

[0032]

[0033] Among them, f ij (x) is the membership degree of the i-th target variable to the j-th fuzzy set, x is the target variable, c ij and b ij are the mean and standard deviation of the membership function of the i-th target variable and the j-th fuzzy set respectively;

[0034] The construction of multiple fuzzy reasoning layers includes the following steps:

[0035] The products of the membership degrees are calculated by using the fuzzy reasoning layers as the outputs of the corresponding fuzzy reasoning layers.

[0036] In some embodiments, the determining of the initial weights and thresholds of the fuzzy neural network based on the radial basis function comprises the following steps:

[0037] The particle swarm optimization algorithm is used to determine the initial weights and thresholds of the fuzzy neural network based on radial basis function.

[0038] In some embodiments, the method of determining the initial weights and thresholds of the fuzzy neural network based on the radial basis function by using the particle swarm optimization algorithm comprises the following steps:

[0039] Initializing the position and velocity of each particle in the search space, and setting a value range of the velocity; wherein the position of each particle represents a set of the initial weights and the thresholds;

[0040] Calculating the fitness value of each particle according to the position and the speed of each particle, and then recording the individual optimal fitness value and the global optimal fitness value of each particle; wherein the global optimal fitness value is the optimal value of the fitness values ​​of all the particles;

[0041] If the current position of each particle obtains a better fitness value, the position and speed of the particle are updated;

[0042] Return to the step of calculating the fitness value of each particle according to the position and the speed of each particle, and then recording the individual optimal fitness value and the global optimal fitness value of each particle, until a preset end condition is reached, and use the position of the particle corresponding to the current global optimal fitness value as the initial weight and the threshold.

[0043] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application provides an anesthesia target control parameter optimization device, the device comprising:

[0044] Model building unit, used to build pharmacokinetic models;

[0045] an anesthesia target control parameter output unit, used to output a first anesthesia target control parameter using the pharmacokinetic model;

[0046] a deviation determination unit, configured to determine a deviation value between the first anesthesia target control parameter and a target anesthesia target control parameter and a rate of change of the deviation value over time as a deviation change rate;

[0047] A network parameter initialization unit, used to determine the initial weights and thresholds of the fuzzy neural network based on the radial basis function;

[0048] A control parameter determination unit, used for inputting the deviation value and the deviation change rate into the radial basis function-based fuzzy neural network provided with the initial weight and the threshold value, to obtain control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters;

[0049] The anesthesia target control parameter optimization unit is used to control the pharmacokinetic model to output a second anesthesia target control parameter based on the control parameter and in combination with the deviation value and the deviation change rate using the PID controller.

[0050] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above method when executing the computer program.

[0051] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0052] The embodiments of the present application include at least the following beneficial effects:

[0053] The present application can construct a pharmacokinetic model; use the pharmacokinetic model to output a first anesthesia target control parameter; determine the deviation value between the first anesthesia target control parameter and the target anesthesia target control parameter and the rate of change of the deviation value over time as the deviation change rate; determine the initial weights and thresholds of the fuzzy neural network based on the radial basis function; input the deviation value and the deviation change rate into the fuzzy neural network based on the radial basis function with the initial weights and thresholds, and obtain the control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters; use the PID controller to control the pharmacokinetic model based on the control parameters and in combination with the deviation value and the deviation change rate to output the second anesthesia target control parameter. The present application can determine the optimized initial weights and thresholds, improve the accuracy of the initial parameter selection of the fuzzy neural network based on the radial basis function, and then obtain the optimized control parameters, enhance the adaptive ability and robustness of the PID controller, and be able to control the pharmacokinetic model to output more accurate anesthesia target control parameters, thereby improving the accuracy of the anesthesia target control parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 A schematic diagram of a flow chart of an anesthesia target control parameter optimization method provided in an embodiment of the present application;

[0056] Figure 2 An example flow chart of anesthesia target control parameter optimization provided in an embodiment of the present application;

[0057] Figure 3 Schematic diagram of the PBPK model provided in the embodiments of the present application;

[0058] Figure 4A schematic diagram of a lung model provided in an embodiment of the present application;

[0059] Figure 5 A schematic diagram of a brain model provided in an embodiment of the present application;

[0060] Figure 6 A simulation diagram of the PBPK model provided in the embodiments of the present application;

[0061] Figure 7 A simulation diagram of a PBPK model including the relationship between BIS and brain drug concentration provided in the embodiments of the present application;

[0062] Figure 8 An example diagram of a fuzzy RBF neural network structure provided in an embodiment of the present application;

[0063] Fig. 9 An example flow chart of a particle swarm algorithm provided in an embodiment of the present application;

[0064] Fig.10 A schematic diagram of the structure of an anesthesia target control parameter optimization device provided in an embodiment of the present application;

[0065] Fig.11 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0067] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0068] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0070] Before describing the embodiments of the present application in detail, some related technologies involved in the embodiments of the present application are first described as follows:

[0071] Existing target-controlled anesthesia infusion systems are mainly based on the classic PID (proportional-integral-differential) control algorithm. This algorithm is simple and easy to implement, but it has some limitations. First, the parameters of the PID controller need to be manually adjusted according to the specific application, which usually requires professional knowledge and experience, and the adjustment process can be time-consuming and imprecise. Secondly, due to the complexity and individual differences of the human physiological system, it is difficult for the PID controller to adapt to the dynamic changes of the patient's physiological state, resulting in inaccurate control of the depth of anesthesia, which may cause risks such as intraoperative awareness or excessive anesthesia. In addition, the traditional PID control algorithm is sensitive to changes in system parameters and external interference, lacks adaptive and learning capabilities, and limits its performance in complex and changeable clinical anesthesia applications.

[0072] At present, the related technology of this application includes an anesthesia target-controlled infusion system that combines fuzzy control and neural network. However, fuzzy control can handle uncertainty and nonlinear problems, while neural network has strong self-learning ability and nonlinear mapping ability. This related technology aims to adjust the parameters of the PID controller through fuzzy logic to adapt to changes in the patient's physiological state, while using neural network to predict and compensate for the dynamic changes of drug effects.

[0073] Although the combination of fuzzy control and neural networks can theoretically improve the adaptability and control accuracy of TCI systems, it still faces some challenges in practical applications. First, there is a lack of systematic methods for the structure and parameter selection of fuzzy neural networks, which often rely on expert experience and trial and error, which limits the generalization ability and real-time performance of the system. Second, the training and optimization of fuzzy neural networks require a large amount of labeled data, which may be difficult to obtain in clinical settings. In addition, the existing fuzzy neural networks have high computational complexity when dealing with high-dimensional data and complex decisions, which may cause system response delays and affect anesthesia safety.

[0074] The present application aims to solve the problems existing in the prior art by proposing an anesthesia target controlled infusion scheme based on an improved fuzzy neural network PID control, which can adaptively adjust the PID parameters so that the output anesthesia target control parameters can adapt to the individual differences and dynamic changes of patients. At the same time, the particle swarm optimization algorithm is used to optimize the initial weights and thresholds of the fuzzy neural network to improve the real-time performance and control accuracy of the present application.

[0075] The embodiment of the present application provides an anesthesia target control parameter optimization method, device, electronic device and storage medium. The technical solution of the present application includes: constructing a pharmacokinetic model; using the pharmacokinetic model to output a first anesthesia target control parameter; determining the deviation value between the first anesthesia target control parameter and the target anesthesia target control parameter and the rate of change of the deviation value over time as the deviation change rate; determining the initial weights and thresholds of the fuzzy neural network based on the radial basis function; inputting the deviation value and the deviation change rate into the fuzzy neural network based on the radial basis function with the initial weights and thresholds, and obtaining the control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters; using the PID controller to control the pharmacokinetic model based on the control parameters and in combination with the deviation value and the deviation change rate to output the second anesthesia target control parameter. The present application can determine the optimized initial weights and thresholds, improve the accuracy of the initial parameter selection of the fuzzy neural network based on the radial basis function, and then obtain the optimized control parameters, enhance the adaptive ability and robustness of the PID controller, and be able to control the pharmacokinetic model to output more accurate anesthesia target control parameters, thereby improving the accuracy of the anesthesia target control parameters.

[0076] The embodiments of the present application provide an anesthesia target control parameter optimization method, device, electronic device and storage medium, which relate to the field of artificial intelligence technology. The anesthesia target control parameter optimization method, device, electronic device and storage medium provided in the embodiments of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms and other basic cloud computing services. The server can also be a node server in a blockchain network; the software can be an application that implements a knowledge extraction method, etc., but is not limited to the above forms.

[0077] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0078] Reference Figure 1 The embodiment of the present application provides an anesthesia target control parameter optimization method, which may include but is not limited to S100 to S150, as follows:

[0079] S100: Build a pharmacokinetic model.

[0080] Further, S100 may include S101:

[0081] S101: Constructing the pharmacokinetic model based on the physiological pharmacodynamic model of propofol in the human body; wherein the physiological pharmacodynamic model of propofol in the human body includes a vein model, a lung model, a brain model, a liver model, a fast equilibrium tissue model, a slow equilibrium tissue model and a deep distribution model.

[0082] More specifically, S101 may include S1011 to S1018:

[0083] S1011: establishing a third-order differential equation of the lung model, and then determining a transfer function expression of the lung model;

[0084] S1012: establishing a second-order differential equation of the brain model, and then determining a transfer function expression of the brain model;

[0085] S1013: establishing a first-order differential equation of the liver model, and then determining a transfer function expression of the liver model;

[0086] S1014: establishing a first-order differential equation of the rapid equilibrium tissue model, and then determining a transfer function expression of the rapid equilibrium tissue model;

[0087] S1015: establishing a second-order differential equation of the slow equilibrium tissue model, and then determining a transfer function expression of the slow equilibrium tissue model;

[0088] S1016: Establishing a first-order differential equation of the vein model, and then determining a transfer function expression of the vein model;

[0089] S1017: Determine a transfer function expression of the deep distribution model;

[0090] S1018: Constructing the pharmacokinetic model according to the transfer function expressions corresponding to the venous model, the lung model, the brain model, the liver model, the fast equilibrium tissue model, the slow equilibrium tissue model and the deep distribution model.

[0091] S110: Outputting a first anesthesia target control parameter using the pharmacokinetic model.

[0092] It can be understood that the first anesthesia target control parameter output by the pharmacokinetic model in this embodiment may not be optimized. Therefore, in order to improve the accuracy and make the anesthesia target control parameter output by the pharmacokinetic model closer to the target anesthesia target control parameter, the present application can continue to execute subsequent steps.

[0093] S120: Determine a deviation value between the first anesthesia target control parameter and a target anesthesia target control parameter and a change rate of the deviation value over time as a deviation change rate.

[0094] It can be understood that the target anesthesia target control parameter is a pre-set target value.

[0095] In some embodiments, before executing S130, the present application may further include the step of constructing a fuzzy neural network based on a radial basis function, which may specifically include:

[0096] Constructing multiple input layers; wherein each of the input layers is used to input multiple target variables, and the target variables include the deviation value and the deviation change rate;

[0097] Constructing a plurality of fuzzification layers; wherein each of the fuzzification layers is used to divide each of the target variables into fuzzy sets equal in number to the fuzzification layers and calculate the membership degree of each of the target variables to each of the fuzzy sets;

[0098] Constructing multiple fuzzy reasoning layers; wherein each of the fuzzy reasoning layers is used to implement fuzzy reasoning according to the output of each of the fuzzification layers;

[0099] Constructing a plurality of output layers; wherein each of the output layers is used to output a target parameter, and the target parameter includes the control parameter;

[0100] The input layers, the fuzzification layers, the fuzzy inference layers and the output layers are connected in sequence to obtain the fuzzy neural network based on radial basis function.

[0101] As a further implementation, constructing multiple fuzzification layers includes the following steps:

[0102] Utilizing each of the fuzzification layers to calculate the membership of each of the target variables to each of the fuzzy sets according to a membership function;

[0103] The membership function is:

[0104]

[0105] Among them, f ij (x) is the membership degree of the i-th target variable to the j-th fuzzy set, x is the target variable, c ij and b ij are the mean and standard deviation of the membership function of the i-th target variable and the j-th fuzzy set respectively.

[0106] As a further implementation, multiple fuzzy reasoning layers are constructed, including the following steps:

[0107] The products of the membership degrees are calculated by using the fuzzy reasoning layers as the outputs of the corresponding fuzzy reasoning layers.

[0108] S130: Determine initial weights and thresholds of the fuzzy neural network based on radial basis functions.

[0109] Further, S130 may include S131:

[0110] S131: Determine the initial weights and thresholds of the fuzzy neural network based on radial basis function using a particle swarm optimization algorithm.

[0111] More specifically, S131 may include S1311 to S1314:

[0112] S1311: Initializing the position and velocity of each particle in the search space, and setting a value range of the velocity; wherein the position of each particle represents a set of the initial weight and the threshold;

[0113] S1312: Calculate the fitness value of each particle according to the position and the speed of each particle, and then record the individual optimal fitness value and the global optimal fitness value of each particle; wherein the global optimal fitness value is the optimal value of the fitness values ​​of all the particles;

[0114] S1313: if the current position of each particle obtains a better fitness value, updating the position and the speed of the particle;

[0115] S1314: Return to the step of calculating the fitness value of each particle according to the position and the speed of each particle, and then recording the individual optimal fitness value and the global optimal fitness value of each particle, until a preset end condition is reached, and use the position of the particle corresponding to the current global optimal fitness value as the initial weight and the threshold.

[0116] S140: Input the deviation value and the deviation change rate into the radial basis function-based fuzzy neural network set with the initial weight and the threshold to obtain control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters.

[0117] S150: Using the PID controller to control the pharmacokinetic model to output a second anesthesia target control parameter based on the control parameter and in combination with the deviation value and the deviation change rate.

[0118] It can be understood that the second anesthesia target control parameter is an optimized anesthesia target control parameter, and the second anesthesia target control parameter obtained after optimization is more accurate and better matches the patient's dynamically changing vital sign parameters.

[0119] Next, the solution of the embodiment of the present application will be introduced and explained in detail with reference to specific application examples.

[0120] Reference Figure 2 , this embodiment provides an example flowchart for anesthesia target control parameter optimization.

[0121] Specifically, this embodiment may include the following contents:

[0122] 1. Construct a pharmacokinetic model.

[0123] Taking the human body as the control object of target-controlled infusion of anesthesia, the changes of drugs entering the human body are dynamic and complex, and there is an interaction between the two. Pharmacodynamics studies the effects of drugs on the human body, including three categories: sedation, analgesia, and muscle relaxation; while pharmacokinetics studies the effects of the human body on drugs, mainly including four categories: drug onset, drug distribution, drug metabolism, and drug elimination. The coupling relationship between the two is inseparable. Only by fully understanding their physiological structure, understanding the dynamic process of anesthetic drugs in the human body, and establishing appropriate pharmacodynamic and pharmacokinetic models can we promote the development of precision anesthesia. At present, drug models are mainly divided into two categories: compartment models and physiological models. The compartment model can accurately and intuitively describe the change process of drug blood concentration over time, but due to the essential differences between the compartment model and the physiological model, the abstractness and subjectivity of its division make the compartment model lack of internal mechanism and anatomical basis. Considering that the impact of anesthetic drugs on patients is hierarchical and organic, it is more appropriate to select the physiological model as the target-controlled infusion model of anesthesia.

[0124] For example, the present embodiment can use the improved human propofol physiological pharmacodynamics model as a pharmacokinetic model (i.e., PBPK model). The PBPK model simplifies the complex internal mechanisms of the human body, highlights key structures, and forms an effective blood drug circulation closed-loop system. The main organs of the human body are highlighted in the PBPK model, including the lungs, brain, liver, veins, fast distribution chambers (fast equilibrium tissues) and slow distribution chambers (slow equilibrium tissues), etc. The schematic diagram of the PBPK model is shown in FIG. Figure 3 shown.

[0125] According to the physiological pharmacodynamics model of propofol, the differential equations of each organ model are established, and the corresponding transfer function expressions are obtained as follows:

[0126] (1) Lung model Figure 4 As shown, the third-order differential equation corresponding to the lung model is as follows:

[0127]

[0128] The transfer function expression of the lung model is:

[0129]

[0130] (2) Brain models such as Figure 5 As shown, the second-order differential equation corresponding to the brain model is as follows:

[0131]

[0132] The transfer function expression of the brain model is:

[0133]

[0134] (3) The first-order differential equation corresponding to the liver model is as follows:

[0135]

[0136] The transfer function expression corresponding to the liver model is:

[0137]

[0138] (4) The first-order differential equation corresponding to the rapid equilibrium organization model is as follows:

[0139]

[0140] The transfer function expression corresponding to the fast equilibrium organization model is:

[0141]

[0142] (5) The second-order differential equation corresponding to the slow equilibrium organization model is as follows:

[0143]

[0144] The transfer function expression corresponding to the slow equilibrium organization model is:

[0145]

[0146] (6) The first-order differential equation corresponding to the vein model is as follows:

[0147]

[0148] The transfer function expression corresponding to the vein model is:

[0149]

[0150] (7) The transfer function expression of the deep distribution model is:

[0151]

[0152] The meanings of the parameters in the above formulas are as follows:

[0153] c pa1 ,c pa2 ,c pa3 Represents the blood drug concentration in the first, second, and third lung chambers respectively; c art is the arterial blood drug concentration; Q co V is cardiac output; p1 ,V p2 ,V p3 E is the volume of the first, second, and third lung compartments;lung is the lung clearance constant; c brn1 ,c brn2 is the blood drug concentration in the first and second compartments of the brain; Q brn V is cerebral blood flow; brn1 ,V brn2 is the volume of the first and second compartments of the brain; PS brn is the product of the brain permeability surface area; c hep is the hepatic blood drug concentration; Q hep V is the liver blood flow; hep is the volume of the liver compartment; E hep is the liver clearance constant; c fast To quickly balance the blood drug concentration in the tissue compartment; Q fast Rapid balance of blood flow to tissues; V fast To quickly balance the volume of tissue compartments; E fast is the elimination rate constant of the fast equilibrium tissue; c slow Q is the blood drug concentration in the slow equilibrium tissue compartment; slow For blood flow to slow balanced tissues; V slow is the volume of the slow equilibrium tissue chamber; PS deep is the permeable surface area of ​​the deep compartment; c deep is the deep chamber blood drug concentration; c v is the drug concentration in venous blood; V mix is the volume of the venous chamber; Q co is cardiac output; c post is the blood drug concentration after injection; Doserate is the injection rate.

[0154] For example, the meaning of the parameters of each model in the PBPK model and the exemplary parameter values ​​are shown in Table 1.

[0155] Table 1

[0156]

[0157]

[0158] By analyzing the PBPK model, substituting the corresponding parameter values ​​into its transfer function, the overall pharmacokinetic modeling is completed with the help of Simulink. In this embodiment, the bispectral index BIS is selected as the anesthesia depth monitoring index, and its reasonable range is set to 40-65. According to the relationship between BIS and brain drug concentration, the reasonable range corresponding to propofol is 2.83mg / L~8.07mg / L. In this embodiment, the anesthesia depth monitoring index BIS is set to 55, and the corresponding propofol concentration is 5mg / L. According to the parameter values ​​in Table 1, they are substituted into the transfer function expressions to obtain the corresponding PBPK model simulation diagram as shown below: Figure 6shown.

[0159] In order to better verify the rationality of the PBPK model, the PBPK model was encapsulated and constructed according to the relationship between BIS and brain drug concentration. Figure 7 The model shown is simulated. Figure 7 This is a simulation diagram of the PBPK model that includes the relationship between BIS and brain drug concentration.

[0160] 2. Establish a fuzzy neural network based on radial basis function (fuzzy RBF neural network).

[0161] Fuzzy control is a computer control technology based on fuzzy mathematics, which includes fuzzy set theory, fuzzy language and fuzzy logic. This intelligent control method is nonlinear, and the processing process fits the human way of thinking and can be cleverly transformed into fuzzy language. Even if it is difficult to establish an accurate model for the controlled object, it can be effectively controlled.

[0162] In the process of designing the fuzzy neural network, PBPK was used as the model basis, and the deviation e and the deviation change rate ec of the collected BIS value and the target BIS value were used as input variables. p , K i , K d As output, a 2-7-7-3 type 4-layer network structure is used but not limited to the fuzzy RBF neural network structure. Figure 8 shown.

[0163] The layers are described as follows:

[0164] (1) Input layer: This layer has two nodes, corresponding to the two input variables: the deviation e between the given BIS value and the actual BIS value, and the deviation change rate ec.

[0165] (2) Fuzzification layer: This layer has 7 nodes, which divide the two inputs into 7 fuzzy sets respectively, and use Gaussian function as the membership function to calculate the degree to which the two input variables belong to the 7 fuzzy sets, namely:

[0166]

[0167] where c ij and b ij are the mean and standard deviation of the Gaussian function of the i input and j fuzzy set.

[0168] (3) Fuzzy reasoning layer. Its function is to implement fuzzy reasoning. In this layer, each node is a fuzzy rule, with a total of 49 fuzzy rules. The output of each node can be calculated as the product of all input signals at the node, that is:

[0169]

[0170] (4) Output layer. This layer has 3 nodes, corresponding to the PID parameters K p , K i , K d The output of each node is a weighted sum of the inputs of the previous layer:

[0171]

[0172] where w ij is the weight connecting the third layer nodes to the output nodes.

[0173] 3. Fuzzy RBF neural network PID controller based on particle swarm optimization.

[0174] In the particle swarm optimization algorithm, each solution in the target space can be represented by a bird (particle), and the optimal solution in the problem is the food source that the flock of birds is looking for. In the process of finding the optimal solution, each particle has individual behavior and group behavior. Each particle will learn from the flight experience of its companions and draw on its own flight experience to find the optimal solution. Each particle will learn from two values, one is the individual's historical optimal value pbest; the other is the group's historical optimal value (global optimal value) gbest. The particle will adjust its own speed and position according to these two values, and the quality of each position is determined by the fitness value. The fitness function is the objective function of the optimization.

[0175] The process of the particle swarm algorithm in this embodiment is as follows Fig. 9 As shown in Figure 2, the basic process of the particle swarm algorithm includes:

[0176] (1) Initialize the particle position X in the search space i and speed V i . Input c1, c2, w, V in sequence max 、V min Among them, w is the initial weight, c1 and c2 are thresholds, V max 、V min are the maximum and minimum speed respectively.

[0177] (2) Calculate the fitness value according to the fitness function and record the individual optimal fitness value p best And the global optimal fitness value g best . Using the fitness function f(X i ) evaluates the fitness value of each particle. The fitness function is the integral of the time-weighted absolute error ITAE:

[0178]

[0179] Where r(t) is the reference signal and y(t) is the system output.

[0180] (3) If the current position has a better fitness value, update the individual best position p of each particle best,i .

[0181] (4) If the individual best position of any particle has a better fitness value, the global best position g is updated best .

[0182] (5) Update the velocity and position of each particle using the following formula:

[0183]

[0184] Where r1 and r2 are random numbers between 0 and 1.

[0185] (6) Check whether the preset end condition (e.g., the maximum number of iterations) is met. If not, return to step (2) of the basic process of the particle swarm algorithm.

[0186] In summary, the technical solution of this embodiment includes:

[0187] 1. Integrated optimization of fuzzy neural network and PID control.

[0188] A novel integrated optimization method is proposed, which combines fuzzy neural network (FNN) with proportional-integral-differential (PID) control. The above integrated method not only takes advantage of the advantages of fuzzy logic in dealing with uncertainty and nonlinear problems, but also combines the stability and robustness of PID control. The real-time adjustment of PID parameters by fuzzy neural network can better adapt to the dynamic changes of the patient's physiological state and improve the flexibility and accuracy of anesthesia target control parameters.

[0189] 2. Application of particle swarm optimization algorithm in fuzzy RBF neural network.

[0190] The particle swarm optimization (PSO) algorithm is applied to the fuzzy radial basis function (RBF) neural network to optimize the initial weights and thresholds of the network. The above optimization method improves the efficiency and effect of network training, so that the fuzzy RBF neural network can converge to the optimal solution faster, thereby improving the overall performance and response speed of the anesthesia target-controlled infusion system.

[0191] 3. Design and implementation of closed-loop anesthesia target-controlled infusion system.

[0192] A closed-loop anesthesia target-controlled infusion system was designed and implemented, which can automatically adjust the infusion rate of anesthetic drugs according to the patient's real-time physiological feedback (such as the bispectral index BIS). The above closed-loop control strategy makes the anesthesia target control parameters more personalized and accurate.

[0193] Since the selection of the initial weights and thresholds of the fuzzy neural network will have a great impact on the tuning effect of the PID parameters, this embodiment uses a particle swarm algorithm to optimize the fuzzy RBF neural network to obtain the optimal initial weights and thresholds, and then uses the optimized fuzzy neural network to tune the various PID parameters to achieve optimal control of anesthesia. The particle swarm optimized fuzzy RBF neural network PID controller combines the optimization ability of the particle swarm and the adaptive adjustment ability of the fuzzy neural network to achieve a better PID parameter tuning effect. The schematic diagram of the TCI system based on the fuzzy RBF neural network PID control based on particle swarm optimization is shown below:

[0194] Compared with the prior art, this embodiment has the following advantages and beneficial effects:

[0195] This embodiment involves a PID control method that combines particle swarm optimization and fuzzy RBF neural network, which has a series of significant advantages and beneficial effects compared to the existing target-controlled infusion control technology. First, this embodiment optimizes the initial weights and thresholds of the fuzzy RBF neural network through the particle swarm optimization algorithm, thereby improving the accuracy of the initial parameter selection of the control system, thereby enhancing the system's adaptability and robustness. The optimization mechanism of this embodiment reduces the reliance on expert experience and trial and error, allowing the system to converge to the optimal solution more quickly, improving control accuracy and response speed.

[0196] Secondly, the control method of this embodiment combines the flexibility of fuzzy logic and the learning ability of neural networks, so that the system can better handle uncertainty and nonlinear problems, especially in the face of dynamic changes and individual differences in the patient's physiological state, and can determine more accurate anesthesia target control parameters. Fuzzy RBF neural networks can simulate complex nonlinear relationships, while PID control provides a stable control framework. The combination of the two enables the system to maintain stability while also having good dynamic tracking capabilities.

[0197] In addition, this embodiment improves the flexibility and adaptability of the system by adjusting the PID parameters in real time, so that the anesthesia target controlled infusion system can determine more accurate anesthesia target controlled parameters.

[0198] In summary, the fuzzy RBF neural network PID control method based on particle swarm optimization has obvious advantages in improving the control performance, safety and adaptability of anesthesia target-controlled infusion system.

[0199] Reference Fig.10 The embodiment of the present application further provides an anesthesia target control parameter optimization device, which can implement the above-mentioned anesthesia target control parameter optimization method, and the device includes:

[0200] Model building unit, used to build pharmacokinetic models;

[0201] an anesthesia target control parameter output unit, used to output a first anesthesia target control parameter using the pharmacokinetic model;

[0202] a deviation determination unit, configured to determine a deviation value between the first anesthesia target control parameter and a target anesthesia target control parameter and a rate of change of the deviation value over time as a deviation change rate;

[0203] A network parameter initialization unit, used to determine the initial weights and thresholds of the fuzzy neural network based on the radial basis function;

[0204] A control parameter determination unit, used for inputting the deviation value and the deviation change rate into the radial basis function-based fuzzy neural network provided with the initial weight and the threshold value, to obtain control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters;

[0205] The anesthesia target control parameter optimization unit is used to control the pharmacokinetic model to output a second anesthesia target control parameter based on the control parameter and in combination with the deviation value and the deviation change rate using the PID controller.

[0206] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0207] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned anesthesia target control parameter optimization method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0208] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0209] See also Fig.11 , Fig.11 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0210] The processor 1101 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0211] The memory 1102 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1102 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1102, and the processor 1101 calls and executes an anesthesia target control parameter optimization method of the embodiment of this application;

[0212] Input / output interface 1103, used to implement information input and output;

[0213] The communication interface 1104 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0214] A bus 1105 that transmits information between various components of the device (e.g., the processor 1101, the memory 1102, the input / output interface 1103, and the communication interface 1104);

[0215] The processor 1101 , the memory 1102 , the input / output interface 1103 and the communication interface 1104 are connected to each other in communication within the device via the bus 1105 .

[0216] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned anesthesia target control parameter optimization method is implemented.

[0217] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0218] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0219] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0220] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0221] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0222] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0223] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0224] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0225] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0226] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0227] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0228] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0229] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for optimizing anesthesia target control parameters, characterized in that: The method comprises the following steps: Constructing pharmacokinetic models; Outputting a first anesthesia target control parameter using the pharmacokinetic model; Determining a deviation value between the first anesthesia target control parameter and a target anesthesia target control parameter and a rate of change of the deviation value over time as a deviation change rate; Determine the initial weights and thresholds of the fuzzy neural network based on radial basis function; Inputting the deviation value and the deviation change rate into the radial basis function-based fuzzy neural network provided with the initial weight and the threshold value to obtain control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters; The PID controller is used to control the pharmacokinetic model to output a second anesthesia target control parameter based on the control parameter and in combination with the deviation value and the deviation change rate.

2. The method for optimizing anesthesia target control parameters according to claim 1, characterized in that: The pharmacokinetic model is constructed, comprising the following steps: The pharmacokinetic model is constructed based on the physiological pharmacodynamic model of propofol in the human body; wherein the physiological pharmacodynamic model of propofol in the human body includes a vein model, a lung model, a brain model, a liver model, a fast equilibrium tissue model, a slow equilibrium tissue model and a deep distribution model.

3. The method for optimizing anesthesia target control parameters according to claim 2, characterized in that: The pharmacokinetic model is constructed based on the physiological pharmacodynamics model of propofol in the human body, comprising the following steps: Establishing a third-order differential equation of the lung model, and then determining a transfer function expression of the lung model; Establishing a second-order differential equation of the brain model, and then determining a transfer function expression of the brain model; Establishing a first-order differential equation of the liver model, and then determining a transfer function expression of the liver model; Establishing a first-order differential equation of the fast equilibrium tissue model, and then determining a transfer function expression of the fast equilibrium tissue model; Establishing a second-order differential equation of the slow equilibrium tissue model, and then determining a transfer function expression of the slow equilibrium tissue model; Establishing a first-order differential equation of the vein model, and then determining a transfer function expression of the vein model; Determining a transfer function expression of the deep distribution model; The pharmacokinetic model is constructed according to the transfer function expressions corresponding to the venous model, the lung model, the brain model, the liver model, the fast equilibrium tissue model, the slow equilibrium tissue model and the deep distribution model.

4. The method for optimizing anesthesia target control parameters according to claim 1, characterized in that: Before determining the initial weights and thresholds of the fuzzy neural network based on the radial basis function, the method further comprises the following steps: Constructing multiple input layers; wherein each of the input layers is used to input multiple target variables, and the target variables include the deviation value and the deviation change rate; Constructing a plurality of fuzzification layers; wherein each of the fuzzification layers is used to divide each of the target variables into fuzzy sets equal in number to the fuzzification layers and calculate the membership degree of each of the target variables to each of the fuzzy sets; Constructing multiple fuzzy reasoning layers; wherein each of the fuzzy reasoning layers is used to implement fuzzy reasoning according to the output of each of the fuzzification layers; Constructing a plurality of output layers; wherein each of the output layers is used to output a target parameter, and the target parameter includes the control parameter; The input layers, the fuzzification layers, the fuzzy inference layers and the output layers are connected in sequence to obtain the fuzzy neural network based on radial basis function.

5. The method for optimizing anesthesia target control parameters according to claim 4, characterized in that: The construction of multiple fuzzy layers includes the following steps: Utilizing each of the fuzzification layers to calculate the membership of each of the target variables to each of the fuzzy sets according to a membership function; The membership function is: Among them, f ij (x) is the membership degree of the i-th target variable to the j-th fuzzy set, x is the target variable, c ij and b ij are the mean and standard deviation of the membership function of the i-th target variable and the j-th fuzzy set respectively; The construction of multiple fuzzy reasoning layers includes the following steps: The products of the membership degrees are calculated by using the fuzzy reasoning layers as the outputs of the corresponding fuzzy reasoning layers.

6. The method for optimizing anesthesia target control parameters according to claim 1, characterized in that: The determination of the initial weights and thresholds of the fuzzy neural network based on the radial basis function comprises the following steps: The particle swarm optimization algorithm is used to determine the initial weights and thresholds of the fuzzy neural network based on radial basis function.

7. The method for optimizing anesthesia target control parameters according to claim 6, characterized in that: The method of using a particle swarm optimization algorithm to determine the initial weights and thresholds of a fuzzy neural network based on a radial basis function comprises the following steps: Initializing the position and velocity of each particle in the search space, and setting a value range of the velocity; wherein the position of each particle represents a set of the initial weights and the thresholds; Calculating the fitness value of each particle according to the position and the speed of each particle, and then recording the individual optimal fitness value and the global optimal fitness value of each particle; wherein the global optimal fitness value is the optimal value of the fitness values ​​of all the particles; If the current position of each particle obtains a better fitness value, the position and speed of the particle are updated; Return to the step of calculating the fitness value of each particle according to the position and the speed of each particle, and then recording the individual optimal fitness value and the global optimal fitness value of each particle, until a preset end condition is reached, and use the position of the particle corresponding to the current global optimal fitness value as the initial weight and the threshold.

8. An anesthesia target control parameter optimization device, characterized in that: The device comprises: Model building unit, used to build pharmacokinetic models; an anesthesia target control parameter output unit, used to output a first anesthesia target control parameter using the pharmacokinetic model; a deviation determination unit, configured to determine a deviation value between the first anesthesia target control parameter and a target anesthesia target control parameter and a rate of change of the deviation value over time as a deviation change rate; A network parameter initialization unit, used to determine the initial weights and thresholds of the fuzzy neural network based on the radial basis function; A control parameter determination unit, used for inputting the deviation value and the deviation change rate into the radial basis function-based fuzzy neural network provided with the initial weight and the threshold value, to obtain control parameters of the PID controller; wherein the control parameters include proportional control parameters, integral control parameters and differential control parameters; The anesthesia target control parameter optimization unit is used to control the pharmacokinetic model to output a second anesthesia target control parameter based on the control parameter and in combination with the deviation value and the deviation change rate using the PID controller.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images