Flow adjusting system and method for anesthesia bomb

Abstract

The invention relates to the technical field of data adjustment, and provides a flow adjustment system and method for an anesthesia bomb, which obtains an optimized flow adjustment model of the anesthesia bomb by obtaining historical adjustment data of the anesthesia bomb, performing preprocessing and feature extraction and then performing training by using a deep learning model. In practical application, the system can receive data of the anesthesia bomb adjusting terminal in real time, compare the data with preset flow adjusting parameters, calculate adjusting errors and optimize model parameters through a genetic algorithm so as to achieve more accurate flow control. The problem that an existing anaesthesia machine flow adjusting mode cannot obtain accurate adjusting control parameters is solved, the accuracy of the anaesthesia process is improved, and the medical operation efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a flow regulation system and method for an anesthetic bullet. Background Art

[0002] The anesthesia machine is a semi-open anesthesia device, which is mainly composed of an anesthesia vaporizer, flow meter, ventilator, breathing circuit and other components. The anesthesia machine delivers anesthetics into the patient's alveoli through a mechanical circuit to form a partial pressure of anesthetic gas. After diffusing into the blood, it inhibits the central nervous system, thereby achieving the effect of general anesthesia. Modern anesthesia machines are also equipped with safety monitoring systems and residual gas removal systems to improve the safety of the anesthesia process.

[0003] The flow regulation of anesthetic bombs refers to the process of controlling the flow of anesthetic gases (including oxygen, helium, etc.) and anesthetic vapors (generated by anesthetic drugs through vaporizers) by adjusting the control panel on the anesthesia machine during the anesthesia process to meet the changes in different anesthesia depths and respiratory needs. The regulation process is important for maintaining the patient's stable anesthesia state.

[0004] The anesthetic cartridge flow regulation system has the following technical pain points in actual use: due to the low regulation accuracy of the anesthesia machine, the limited adjustable range of the anesthetic cartridge flow, and the lack of effective data collection and processing methods, the control parameters in the flow regulation process cannot achieve the expected flow regulation requirements, resulting in the existing anesthesia machine flow regulation method being unable to obtain accurate regulation control parameters according to the regulation requirements. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a flow regulation system and method for an anesthetic bomb, which solves the problem that the existing flow regulation method of an anesthesia machine cannot obtain accurate regulation control parameters according to regulation requirements.

[0006] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows: In a first aspect, the present invention provides a flow regulation system for an anesthetic bomb, comprising: A data acquisition unit, used to acquire historical data of anesthetic bomb adjustment, the historical data of anesthetic bomb adjustment includes historical data of anesthetic bomb adjustment flow rate, historical data of pressure, historical data of concentration, historical data of time, historical data of system status and historical data of patient physiology; A model building unit is used to preprocess the historical data of anesthetic bullet adjustment to obtain the preprocessed historical data of anesthetic bullet adjustment, extract features from the historical data of anesthetic bullet adjustment, use the feature data of anesthetic bullet adjustment, and train a deep learning model using the feature data of anesthetic bullet adjustment to obtain a flow regulation model of anesthetic bullet; A task receiving unit is used to receive a flow regulation task of an anesthetic bomb, wherein the flow regulation task of the anesthetic bomb includes a target flow regulation ratio, and the target flow regulation ratio is substituted into a flow regulation model of the anesthetic bomb, and the flow regulation model of the anesthetic bomb outputs a flow regulation parameter; The adjustment parameter comparison unit is used to send the flow adjustment parameter to the anesthetic bomb adjustment terminal, collect the real-time data of the anesthetic bomb adjustment terminal, compare the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameter, and obtain real-time adjustment error data; The adjustment parameter optimization unit is used to optimize the flow adjustment model of the anesthetic bullet through a genetic algorithm using the real-time adjustment error data and the historical adjustment data of the anesthetic bullet to obtain the optimized flow adjustment model of the anesthetic bullet, and use the optimized flow adjustment model of the anesthetic bullet to perform data processing on the target flow adjustment ratio to obtain the optimized flow adjustment parameter.

[0007] Furthermore, in the flow regulating system of the anesthetic bomb of the present invention, the model building unit is also used for: In the model building unit, the specific steps of building the flow regulation model of the anesthetic bullet are as follows: preprocessing the acquired historical data of anesthetic bullet regulation, the preprocessing process includes performing duplicate data detection on the historical data of anesthetic bullet regulation, deduplicating the duplicate data, and obtaining the preprocessed historical data of anesthetic bullet regulation; The preprocessed anesthetic bullet adjustment historical data is subjected to feature extraction to obtain anesthetic bullet adjustment feature data, which includes anesthetic bullet adjustment flow feature, anesthetic gas pressure feature, anesthetic gas concentration feature and anesthetic time feature. The extracted anesthetic bullet adjustment feature data is used to train a deep learning model to obtain an anesthetic bullet flow adjustment model.

[0008] Furthermore, in the flow regulating system of the anesthetic bomb of the present invention, the regulating parameter comparison unit is further used for: The real-time data of the anesthetic bomb regulating terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration; The flow adjustment parameters include a desired anesthetic gas flow rate, a desired anesthetic gas pressure, and a desired anesthetic gas concentration; Compare the real-time data of the anesthetic bomb regulating terminal with the flow regulating parameters, including the comparison between the real-time anesthetic gas flow and the expected anesthetic gas flow, the comparison between the real-time anesthetic gas pressure and the expected anesthetic gas pressure, and the comparison between the real-time anesthetic gas concentration and the expected anesthetic gas concentration; The difference between the real-time value and the expected value of each indicator in the process of data comparison between the real-time data of the anesthetic bomb adjustment terminal and the flow adjustment parameter is calculated to obtain the adjustment error data. The adjustment error data includes the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

[0009] Furthermore, in the flow regulating system of the anesthetic bomb of the present invention, the regulating optimization unit is also used for: Integrate the real-time adjustment error data with the anesthetic bomb adjustment history data to obtain an adjustment data set to be processed, wherein the processed adjustment data set includes anesthetic bomb adjustment flow history data, pressure history data, concentration history data, time history data, system state history data, patient physiological history data and real-time adjustment error data; Receive and set the initial parameters of the genetic algorithm, including the population size, genetic generation, crossover probability, and mutation probability of the genetic algorithm; An initial population is created, in which each individual represents a potential parameter set of the anesthetic bullet flow regulation model, and the parameter set of the anesthetic bullet flow regulation model is randomly generated; Fitness evaluation, for each individual in the population, each individual is each parameter set, each parameter set is evaluated using the to-be-processed adjustment data set, the evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb, and calculating the error between the model output and the actual regulation demand, and assigning a fitness value to each individual according to the error between the model output and the actual regulation demand; Selection operation, which selects a preset number of individuals as parents from the population through the tournament selection algorithm according to the fitness value; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation: perform mutation operation on offspring individuals to generate new individuals; Merge the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population; Repeat the selection operation, crossover operation and mutation operation until the predetermined genetic generation number is obtained or the minimum error range is met, and then stop the iterative optimization; After the iterative optimization process is completed, the individual with the highest fitness value is selected from the final population. The individual with the highest fitness value is the optimal parameter set, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model.

[0010] In a second aspect, the present invention provides a flow rate regulating method for an anesthetic bomb, which is applied to the flow rate regulating system of the anesthetic bomb as described above, comprising: Step S101, obtaining historical data of anesthetic bomb adjustment, the historical data of anesthetic bomb adjustment including historical data of anesthetic bomb adjustment flow rate, historical data of pressure, historical data of concentration, historical data of time, historical data of system status and historical data of patient physiology; Step S102, preprocessing the anesthetic bullet adjustment history data to obtain the preprocessed anesthetic bullet adjustment history data, extracting features from the anesthetic bullet adjustment history data, using the anesthetic bullet adjustment feature data, and training a deep learning model using the anesthetic bullet adjustment feature data to obtain a flow regulation model for the anesthetic bullet; Step S103, receiving an anesthetic bomb flow regulation task, the anesthetic bomb flow regulation task including a target flow regulation ratio, substituting the target flow regulation ratio into a flow regulation model of the anesthetic bomb, and the flow regulation model of the anesthetic bomb outputs a flow regulation parameter; Step S104, sending the flow adjustment parameter to the anesthetic bomb adjustment terminal, collecting real-time data of the anesthetic bomb adjustment terminal, comparing the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameter, and obtaining real-time adjustment error data; Step S105, optimizing the flow regulation model of the anesthetic bomb by using the real-time adjustment error data and the historical adjustment data of the anesthetic bomb through a genetic algorithm to obtain an optimized flow regulation model of the anesthetic bomb, and using the optimized flow regulation model of the anesthetic bomb to perform data processing on the target flow regulation ratio to obtain an optimized flow regulation parameter.

[0011] Furthermore, in the flow regulating system of the anesthetic bomb of the present invention, the step S102 comprises: In the model building unit, the specific steps of building the flow regulation model of the anesthetic bullet are as follows: preprocessing the acquired historical data of anesthetic bullet regulation, the preprocessing process includes performing duplicate data detection on the historical data of anesthetic bullet regulation, deduplicating the duplicate data, and obtaining the preprocessed historical data of anesthetic bullet regulation; The preprocessed anesthetic bullet adjustment historical data is subjected to feature extraction to obtain anesthetic bullet adjustment feature data, which includes anesthetic bullet adjustment flow feature, anesthetic gas pressure feature, anesthetic gas concentration feature and anesthetic time feature. The extracted anesthetic bullet adjustment feature data is used to train a deep learning model to obtain an anesthetic bullet flow adjustment model.

[0012] Furthermore, in the flow regulating system of the anesthetic bomb of the present invention, the step S104 comprises: The real-time data of the anesthetic bomb regulating terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration; The flow adjustment parameters include a desired anesthetic gas flow rate, a desired anesthetic gas pressure, and a desired anesthetic gas concentration; Compare the real-time data of the anesthetic bomb regulating terminal with the flow regulating parameters, including the comparison between the real-time anesthetic gas flow and the expected anesthetic gas flow, the comparison between the real-time anesthetic gas pressure and the expected anesthetic gas pressure, and the comparison between the real-time anesthetic gas concentration and the expected anesthetic gas concentration; The difference between the real-time value and the expected value of each indicator in the process of data comparison between the real-time data of the anesthetic bomb adjustment terminal and the flow adjustment parameter is calculated to obtain the adjustment error data. The adjustment error data includes the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

[0013] Furthermore, in the flow regulating system of the anesthetic bomb of the present invention, the step S105 comprises: Integrate the real-time adjustment error data with the anesthetic bomb adjustment history data to obtain an adjustment data set to be processed, wherein the processed adjustment data set includes anesthetic bomb adjustment flow history data, pressure history data, concentration history data, time history data, system state history data, patient physiological history data and real-time adjustment error data; Receive and set the initial parameters of the genetic algorithm, including the population size, genetic generation, crossover probability, and mutation probability of the genetic algorithm; An initial population is created, in which each individual represents a potential parameter set of the anesthetic bullet flow regulation model, and the parameter set of the anesthetic bullet flow regulation model is randomly generated; Fitness evaluation, for each individual in the population, each individual is each parameter set, each parameter set is evaluated using the to-be-processed adjustment data set, the evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb, and calculating the error between the model output and the actual regulation demand, and assigning a fitness value to each individual according to the error between the model output and the actual regulation demand; Selection operation, which selects a preset number of individuals as parents from the population through the tournament selection algorithm according to the fitness value; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation: perform mutation operation on offspring individuals to generate new individuals; Merge the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population; Repeat the selection operation, crossover operation and mutation operation until the predetermined genetic generation number is obtained or the minimum error range is met, and then stop the iterative optimization; After the iterative optimization process is completed, the individual with the highest fitness value is selected from the final population. The individual with the highest fitness value is the optimal parameter set, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model.

[0014] Beneficial effects of the present invention: By acquiring historical data of anesthetic bullet regulation and preprocessing and feature extraction of these data, the present invention can fully utilize historical experience to train a deep learning model. This enables the model to learn the complex laws of anesthetic bullet flow regulation, thereby more accurately controlling the flow of anesthetic bullets in practical applications.

[0015] The present invention can ensure that the parameters such as the flow rate, pressure and concentration of anesthetic gas are always controlled within the ideal range by optimizing the flow rate regulation model of the anesthetic bomb, thereby reducing the risk during the anesthesia process and improving the safety of patients. Anesthesia is an indispensable part of medical surgery, and the flow rate regulation of the anesthetic bomb directly affects the anesthetic effect. The present invention provides more accurate flow rate regulation parameters, allowing anesthesiologists to better control the depth and time of anesthesia, thereby improving the efficiency of the operation.

[0016] The existing flow rate regulation methods of anesthesia machines often rely on experience or simple algorithms, and cannot obtain accurate regulation control parameters. The present invention breaks this limitation by introducing data regulation and machine learning methods, and provides a more scientific and accurate method for regulating the flow rate of anesthesia bombs.

[0017] In summary, the present invention provides an anesthetic bomb flow regulation system and method based on data regulation and machine learning, which not only improves the accuracy of anesthetic bomb flow regulation, but also improves the efficiency of medical surgery, and solves the shortcomings of existing anesthesia machine flow regulation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0019] Figure 1 A schematic flow chart of a flow rate regulation method for an anesthetic bomb provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0021] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0022] In a first aspect, the present invention provides a flow regulation system for an anesthetic bomb, comprising: A data acquisition unit, used to acquire historical data of anesthetic bomb adjustment, the historical data of anesthetic bomb adjustment includes historical data of anesthetic bomb adjustment flow rate, historical data of pressure, historical data of concentration, historical data of time, historical data of system status and historical data of patient physiology; The data acquisition unit collects various historical data in the anesthetic bullet flow regulation system through sensors and monitoring equipment during the anesthetic bullet regulation process, including anesthetic bullet regulation flow, pressure, concentration, time, system status, and patient physiological data. The sensors and monitoring equipment can capture this data in real time and transmit it to the data acquisition unit.

[0023] After receiving the data, the data acquisition unit will perform preliminary processing and organization on it, and then store it in the system. The stored data will provide an important basis for subsequent data preprocessing and model building. By analyzing and mining these data, a more accurate anesthetic bomb flow regulation model can be constructed, thereby improving the accuracy of the anesthesia process.

[0024] A model building unit is used to preprocess the historical data of anesthetic bullet adjustment to obtain the preprocessed historical data of anesthetic bullet adjustment, extract features from the historical data of anesthetic bullet adjustment, use the feature data of anesthetic bullet adjustment, and train a deep learning model using the feature data of anesthetic bullet adjustment to obtain a flow regulation model of anesthetic bullet; The model building unit is responsible for preprocessing and feature extraction of the historical data of anesthetic bullet regulation in the anesthetic bullet flow regulation system, and uses these feature data to train the deep learning model, and finally obtains the flow regulation model of the anesthetic bullet. The following is a detailed description of the steps: Data preprocessing, duplicate data detection and deduplication: The model building unit first performs duplicate data detection on the acquired anesthesia bomb adjustment history data and deduplication the duplicate data. This step can ensure the uniqueness and accuracy of the data and provide a high-quality data foundation for subsequent feature extraction and model training.

[0025] Feature data extraction: Extract features from the preprocessed anesthesia bomb regulation history data to obtain anesthesia bomb regulation feature data. These feature data include anesthesia bomb regulation flow characteristics, anesthesia gas pressure characteristics, anesthesia gas concentration characteristics, and anesthesia time characteristics. These features can reflect the changes in important parameters during anesthesia and are crucial to the construction of a flow regulation model.

[0026] Deep learning model training: The deep learning model is trained using the extracted anesthetic bullet adjustment feature data. By iteratively optimizing the model parameters, the model can learn the complex relationship between the features and the flow adjustment parameters. After the training is completed, the model can output the corresponding flow adjustment parameters based on the input anesthetic bullet adjustment feature data, thereby having the ability to predict and adjust the anesthetic bullet flow.

[0027] Together, they constitute the core function of the model building unit, ensuring the accuracy and reliability of the anesthetic bullet flow regulation system. Through data preprocessing, feature extraction, and deep learning model training, the model building unit can build an efficient and accurate anesthetic bullet flow regulation model, providing strong support for the precise control of the anesthesia process.

[0028] A task receiving unit is used to receive a flow regulation task of an anesthetic bomb, wherein the flow regulation task of the anesthetic bomb includes a target flow regulation ratio, and the target flow regulation ratio is substituted into a flow regulation model of the anesthetic bomb, and the flow regulation model of the anesthetic bomb outputs a flow regulation parameter; The task receiving unit receives the anesthetic bomb flow regulation task in the anesthetic bomb flow regulation system, including the target flow regulation ratio, and substitutes it into the flow regulation model of the anesthetic bomb to output the flow regulation parameters. The following is a detailed process description: Receiving anesthetic bomb flow regulation tasks: The task receiving unit receives anesthetic bomb flow regulation tasks from users or other system components. These tasks usually include specific target flow regulation ratios, that is, the anesthetic gas flow ratio that the user or system hopes to achieve.

[0029] Processing the target flow rate: After receiving the target flow rate, the task receiving unit will process it into a recognizable input format. This step ensures that the flow rate model of the anesthetic bomb can correctly understand and process this input.

[0030] Substituting into the flow regulation model of the anesthetic bomb: The task receiving unit substitutes the processed target flow regulation ratio into the pre-trained flow regulation model of the anesthetic bomb. This model is built based on deep learning technology and can output corresponding flow regulation parameters according to the input target flow regulation ratio.

[0031] Output flow regulation parameters: After the model processes the input target flow regulation ratio, it will output the corresponding flow regulation parameters. These parameters may include the expected anesthetic gas flow, pressure, concentration, etc., which are used to guide the anesthetic bomb regulation terminal to perform actual flow regulation operations.

[0032] Parameter transfer: Finally, the task receiving unit transfers the flow adjustment parameters output by the model to the adjustment parameter comparison unit for subsequent real-time data comparison and optimization processing.

[0033] The task receiving unit enables the anesthetic bomb flow regulation system to automatically calculate appropriate flow regulation parameters according to the actual needs of the user, thereby achieving precise anesthetic gas flow control.

[0034] The adjustment parameter comparison unit is used to send the flow adjustment parameter to the anesthetic bomb adjustment terminal, collect the real-time data of the anesthetic bomb adjustment terminal, compare the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameter, and obtain real-time adjustment error data; The adjustment parameter comparison unit in the anesthetic bomb flow regulation system sends the flow regulation parameters to the anesthetic bomb regulation terminal, collects the real-time data of the terminal, and compares these data with the flow regulation parameters to obtain real-time adjustment error data. The following is a detailed process description: The adjustment parameter comparison unit sends the flow adjustment parameters output by the flow adjustment model of the anesthetic bomb to the anesthetic bomb adjustment terminal. These parameters may include the expected anesthetic gas flow, the expected anesthetic gas pressure and the expected anesthetic gas concentration.

[0035] The adjustment parameter comparison unit collects real-time data from the anesthetic bomb adjustment terminal, including real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration.

[0036] The adjustment parameter comparison unit compares the collected real-time data with the sent flow adjustment parameters. The comparison process includes comparing the real-time anesthetic gas flow rate with the expected anesthetic gas flow rate, comparing the real-time anesthetic gas pressure with the expected anesthetic gas pressure, and comparing the real-time anesthetic gas concentration with the expected anesthetic gas concentration.

[0037] The adjustment parameter comparison unit calculates the difference between the real-time data of the anesthetic bomb adjustment terminal and the expected value of each indicator during the comparison process of the flow adjustment parameter, and obtains the adjustment error data. These data may show that the actual value is higher than the expected value, the actual value is lower than the expected value, or the actual value is consistent with the expected value.

[0038] The adjustment parameter comparison unit can accurately evaluate the difference between the real-time status of the anesthetic bomb adjustment terminal and the expected adjustment target, and provide necessary input for subsequent optimization.

[0039] The adjustment parameter optimization unit is used to optimize the flow adjustment model of the anesthetic bullet through a genetic algorithm using the real-time adjustment error data and the historical adjustment data of the anesthetic bullet to obtain the optimized flow adjustment model of the anesthetic bullet, and use the optimized flow adjustment model of the anesthetic bullet to perform data processing on the target flow adjustment ratio to obtain the optimized flow adjustment parameter.

[0040] The adjustment parameter optimization unit is responsible for optimizing the flow regulation model of the anesthetic bullet by using the real-time adjustment error data and the historical adjustment data of the anesthetic bullet through the genetic algorithm in the anesthetic bullet flow regulation system, and obtaining the optimized model parameters. The following is a detailed process description: The adjustment parameter optimization unit first integrates the real-time adjustment error data with the historical data of the anesthetic bomb adjustment to obtain an adjustment data set to be processed. This data set covers the flow, pressure, concentration, time, system status, patient physiological data and real-time adjustment error data of the anesthetic bomb adjustment.

[0041] The parameter optimization unit receives and sets the initial parameters of the genetic algorithm, including the population size, genetic generation, crossover probability, and mutation probability of the genetic algorithm. The setting of these parameters is crucial to the performance and effect of the genetic algorithm.

[0042] Create an initial population, adjust the parameter optimization unit to create an initial population, in which each individual represents a potential anesthetic bullet flow regulation model parameter set. These parameter sets are randomly generated and provide an initial search space for the subsequent genetic algorithm.

[0043] Fitness evaluation,For each individual in the population (i.e., each parameter set), the unit evaluates it using the regulation data set to be processed. The evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb and calculating the error between the model output and the actual regulation requirement. Based on this error, a fitness value is assigned to each individual to reflect its degree of excellence.

[0044] Selection operation, the adjustment parameter optimization unit selects a preset number of individuals as parents from the population through the tournament selection algorithm according to the fitness value. These parent individuals have higher fitness values ​​and are considered to be better solutions.

[0045] Crossover operation, which performs a crossover operation on the selected parent individuals to generate new offspring individuals. The crossover operation generates new solutions by combining the genes (i.e., parameters) of the parent individuals, which helps to explore new areas in the search space.

[0046] Mutation operation, which performs mutation operation on offspring individuals to generate new individuals. Mutation operation introduces diversity by randomly changing the genes (i.e. parameters) of individuals, which helps to prevent the algorithm from falling into the local optimal solution.

[0047] A new population is formed by merging the offspring individuals generated after the crossover and mutation operations with the remaining unselected parent individuals to form a new population. This new population will serve as the starting point for the next round of genetic algorithm.

[0048] Iterative optimization: Repeat the selection, crossover, and mutation operations until the predetermined number of genetic generations is reached or the minimum error range is met, and then stop iterative optimization. In this process, the individuals in the population will gradually evolve into a solution that is more adapted to the environment.

[0049] Select the optimal parameter set. After the iterative optimization process is completed, select the individual with the highest fitness value from the final population, that is, the optimal parameter set. This optimal parameter set represents the best parameter configuration of the anesthetic bomb flow regulation model.

[0050] The model parameters are updated, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model. This step completes the parameter update of the model, so that the model can more accurately reflect the flow regulation characteristics of the anesthetic bullet.

[0051] Process the target flow rate adjustment ratio, use the optimized flow rate adjustment model of the anesthetic bomb to process the target flow rate adjustment ratio, and obtain the optimized flow rate adjustment parameters. These parameters will be used to guide the anesthetic bomb adjustment terminal to perform actual flow adjustment operations.

[0052] The adjustment parameter optimization unit can gradually adjust the parameters of the anesthetic bomb flow regulation model to improve the accuracy and reliability of the model, thereby more accurately meeting the requirements of the target flow regulation ratio.

[0053] Specifically, the flow regulating system of the anesthetic bomb of the present invention, the model building unit is also used for: In the model building unit, the specific steps of building the flow regulation model of the anesthetic bullet are as follows: preprocessing the acquired historical data of anesthetic bullet regulation, the preprocessing process includes performing duplicate data detection on the historical data of anesthetic bullet regulation, deduplicating the duplicate data, and obtaining the preprocessed historical data of anesthetic bullet regulation; The preprocessed anesthetic bullet adjustment historical data is subjected to feature extraction to obtain anesthetic bullet adjustment feature data, which includes anesthetic bullet adjustment flow feature, anesthetic gas pressure feature, anesthetic gas concentration feature and anesthetic time feature. The extracted anesthetic bullet adjustment feature data is used to train a deep learning model to obtain an anesthetic bullet flow adjustment model.

[0054] In the flow regulation system of the anesthetic bullet, the model building unit is responsible for building the flow regulation model of the anesthetic bullet. This process includes the following specific steps: The acquired anesthesia bullet adjustment history data is preprocessed. This preprocessing process includes duplicate detection of the data and deduplication of the duplicate data, thereby obtaining the preprocessed anesthesia bullet adjustment history data. This step ensures the uniqueness and accuracy of the data, and provides a high-quality data foundation for subsequent feature extraction and model training.

[0055] The feature extraction of the preprocessed anesthesia bomb regulation history data is performed to obtain the anesthesia bomb regulation feature data. These feature data cover the anesthesia bomb regulation flow characteristics, anesthesia gas pressure characteristics, anesthesia gas concentration characteristics and anesthesia time characteristics. These features can fully reflect the changes in important parameters during anesthesia and are crucial for the construction of a flow regulation model.

[0056] The deep learning model is trained using the extracted anesthetic bullet regulation feature data. By iteratively optimizing the model parameters, the model can learn the complex relationship between the features and the flow regulation parameters. After the training is completed, the model can output the corresponding flow regulation parameters based on the input anesthetic bullet regulation feature data, thereby having the ability to predict and regulate the flow of anesthetic bullets.

[0057] In summary, the model building unit constructs an efficient and accurate anesthetic bomb flow regulation model through three key steps: data preprocessing, feature extraction, and deep learning model training. This model provides strong support for the precise control of the anesthesia process.

[0058] Specifically, in the flow regulating system of the anesthetic bomb of the present invention, the regulating parameter comparison unit is further used for: The real-time data of the anesthetic bomb regulating terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration; The flow adjustment parameters include a desired anesthetic gas flow rate, a desired anesthetic gas pressure, and a desired anesthetic gas concentration; Compare the real-time data of the anesthetic bomb regulating terminal with the flow regulating parameters, including the comparison between the real-time anesthetic gas flow and the expected anesthetic gas flow, the comparison between the real-time anesthetic gas pressure and the expected anesthetic gas pressure, and the comparison between the real-time anesthetic gas concentration and the expected anesthetic gas concentration; The difference between the real-time value and the expected value of each indicator in the process of data comparison between the real-time data of the anesthetic bomb adjustment terminal and the flow adjustment parameter is calculated to obtain the adjustment error data. The adjustment error data includes the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

[0059] The adjustment parameter comparison unit is responsible for comparing the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameters in the anesthetic bomb flow adjustment system, and calculating the adjustment error data. Specifically, this process includes the following detailed steps: The real-time data of the anesthetic bomb adjustment terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration, which reflects the actual situation during the anesthesia process and is the basis for subsequent comparison and error calculation.

[0060] The flow regulation parameters include the desired anesthetic gas flow rate, the desired anesthetic gas pressure, and the desired anesthetic gas concentration. These parameters are preset according to the anesthesia requirements and represent the ideal anesthesia state.

[0061] The real-time data of the anesthetic bomb regulation terminal is compared with the flow regulation parameters. The comparison process covers the comparison of the real-time anesthetic gas flow rate with the expected anesthetic gas flow rate, the comparison of the real-time anesthetic gas pressure with the expected anesthetic gas pressure, and the comparison of the real-time anesthetic gas concentration with the expected anesthetic gas concentration, aiming to evaluate the difference between the actual anesthetic state and the ideal state.

[0062] The difference between the real-time value and the expected value of each indicator is calculated to obtain the adjustment error data. The adjustment error data reflects the degree of deviation between the actual anesthesia state and the ideal state, including the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

[0063] Through these detailed steps, the adjustment parameter comparison unit can accurately evaluate the difference between the real-time status during anesthesia and the expected target, providing the necessary basis for subsequent optimization and adjustment.

[0064] Specifically, the flow regulation system of the anesthetic bomb of the present invention, the regulation optimization unit is also used for: Integrate the real-time adjustment error data with the anesthetic bomb adjustment history data to obtain an adjustment data set to be processed, wherein the processed adjustment data set includes anesthetic bomb adjustment flow history data, pressure history data, concentration history data, time history data, system state history data, patient physiological history data and real-time adjustment error data; Receive and set the initial parameters of the genetic algorithm, including the population size, genetic generation, crossover probability, and mutation probability of the genetic algorithm; An initial population is created, in which each individual represents a potential parameter set of the anesthetic bullet flow regulation model, and the parameter set of the anesthetic bullet flow regulation model is randomly generated; Fitness evaluation, for each individual in the population, each individual is each parameter set, each parameter set is evaluated using the to-be-processed adjustment data set, the evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb, and calculating the error between the model output and the actual regulation demand, and assigning a fitness value to each individual according to the error between the model output and the actual regulation demand; Selection operation, which selects a preset number of individuals as parents from the population through the tournament selection algorithm according to the fitness value; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation: perform mutation operation on offspring individuals to generate new individuals; Merge the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population; Repeat the selection operation, crossover operation and mutation operation until the predetermined genetic generation number is obtained or the minimum error range is met, and then stop the iterative optimization; After the iterative optimization process is completed, the individual with the highest fitness value is selected from the final population. The individual with the highest fitness value is the optimal parameter set, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model.

[0065] The regulation optimization unit uses genetic algorithm to optimize the flow regulation model of the anesthetic bomb in the flow regulation system of the anesthetic bomb. The following are the specific steps performed by the unit: The real-time adjustment error data is integrated with the historical data of anesthetic bullet adjustment to form an adjustment data set to be processed, which contains all-round information such as the flow, pressure, concentration, time, system status, patient physiological data and real-time adjustment error of the anesthetic bullet adjustment.

[0066] Genetic algorithm parameter setting, receive and set the initial parameters of the genetic algorithm, which include population size (i.e. the number of potential solutions considered simultaneously), genetic generations (i.e. the number of iterations), crossover probability (i.e. the probability that two solutions merge to produce a new solution), and mutation probability (i.e. the probability that a solution undergoes random changes).

[0067] Initial population creation,Create an initial population, where each individual represents a potential set of parameters for the anesthetic bullet flow regulation model. These parameter sets are randomly generated and provide an initial search space for the genetic algorithm.

[0068] Fitness evaluation, the fitness of each individual in the population (i.e. each parameter set) is evaluated. The evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb and calculating the error between the model output and the actual regulation demand. Based on this error, a fitness value is assigned to each individual to reflect its quality.

[0069] The selection operation selects a preset number of individuals from the population as parents through the tournament selection algorithm based on the fitness value. These parent individuals have higher fitness values ​​and are considered to be better solutions.

[0070] Crossover operation, which performs a crossover operation on the selected parent individuals to generate new offspring individuals. The crossover operation generates new solutions by combining the genes (i.e., parameters) of the parent individuals, which helps to explore new areas in the search space.

[0071] Mutation operation, which performs mutation operation on offspring individuals to generate new individuals. Mutation operation introduces diversity by randomly changing the genes (i.e. parameters) of individuals, which helps to prevent the algorithm from falling into the local optimal solution.

[0072] A new population is formed by merging the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population. This new population will serve as the starting point for the next round of genetic algorithm.

[0073] Iterative optimization, repeating the selection operation, crossover operation and mutation operation until the predetermined genetic generation is reached or the minimum error range is met, and then the iterative optimization is stopped. In this process, the individuals in the population will gradually evolve into a solution that is more adapted to the environment.

[0074] Select the optimal parameter set. After the iterative optimization process is completed, select the individual with the highest fitness value from the final population, that is, the optimal parameter set. The optimal parameter set represents the best parameter configuration of the anesthetic bullet flow regulation model.

[0075] The model parameters are updated, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model. This step completes the parameter update of the model, so that the model can more accurately reflect the flow regulation characteristics of the anesthetic bullet and output more accurate flow regulation parameters.

[0076] The regulation optimization unit can use genetic algorithms to efficiently optimize the anesthetic bomb flow regulation model, thereby improving the accuracy and safety of the anesthesia process.

[0077] In a second aspect, the present invention provides a flow rate regulation method for an anesthetic bomb, which is applied to the flow rate regulation system of the anesthetic bomb as described above, comprising: Step S101, obtaining historical data of anesthetic bomb adjustment, the historical data of anesthetic bomb adjustment including historical data of anesthetic bomb adjustment flow rate, historical data of pressure, historical data of concentration, historical data of time, historical data of system status and historical data of patient physiology; Step S102, preprocessing the anesthetic bullet adjustment history data to obtain the preprocessed anesthetic bullet adjustment history data, extracting features from the anesthetic bullet adjustment history data, using the anesthetic bullet adjustment feature data, and training a deep learning model using the anesthetic bullet adjustment feature data to obtain a flow regulation model for the anesthetic bullet; Step S103, receiving an anesthetic bomb flow regulation task, the anesthetic bomb flow regulation task including a target flow regulation ratio, substituting the target flow regulation ratio into a flow regulation model of the anesthetic bomb, and the flow regulation model of the anesthetic bomb outputs a flow regulation parameter; Step S104, sending the flow adjustment parameter to the anesthetic bomb adjustment terminal, collecting real-time data of the anesthetic bomb adjustment terminal, comparing the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameter, and obtaining real-time adjustment error data; Step S105, optimizing the flow regulation model of the anesthetic bomb by using the real-time adjustment error data and the historical adjustment data of the anesthetic bomb through a genetic algorithm to obtain an optimized flow regulation model of the anesthetic bomb, and using the optimized flow regulation model of the anesthetic bomb to perform data processing on the target flow regulation ratio to obtain an optimized flow regulation parameter.

[0078] Specifically, the flow regulating system of the anesthetic bomb of the present invention, the step S102 includes: In the model building unit, the specific steps of building the flow regulation model of the anesthetic bullet are as follows: preprocessing the acquired historical data of anesthetic bullet regulation, the preprocessing process includes performing duplicate data detection on the historical data of anesthetic bullet regulation, deduplicating the duplicate data, and obtaining the preprocessed historical data of anesthetic bullet regulation; The preprocessed anesthetic bullet adjustment historical data is subjected to feature extraction to obtain anesthetic bullet adjustment feature data, which includes anesthetic bullet adjustment flow feature, anesthetic gas pressure feature, anesthetic gas concentration feature and anesthetic time feature. The extracted anesthetic bullet adjustment feature data is used to train a deep learning model to obtain an anesthetic bullet flow adjustment model.

[0079] Specifically, the flow regulating system of the anesthetic bomb of the present invention, the step S104 includes: The real-time data of the anesthetic bomb regulating terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration; The flow adjustment parameters include a desired anesthetic gas flow rate, a desired anesthetic gas pressure, and a desired anesthetic gas concentration; Compare the real-time data of the anesthetic bomb regulating terminal with the flow regulating parameters, including the comparison between the real-time anesthetic gas flow and the expected anesthetic gas flow, the comparison between the real-time anesthetic gas pressure and the expected anesthetic gas pressure, and the comparison between the real-time anesthetic gas concentration and the expected anesthetic gas concentration; The difference between the real-time value and the expected value of each indicator in the process of data comparison between the real-time data of the anesthetic bomb adjustment terminal and the flow adjustment parameter is calculated to obtain the adjustment error data. The adjustment error data includes the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

[0080] Specifically, the flow regulating system of the anesthetic bomb of the present invention, the step S105 includes: Integrate the real-time adjustment error data with the anesthetic bomb adjustment history data to obtain an adjustment data set to be processed, wherein the processed adjustment data set includes anesthetic bomb adjustment flow history data, pressure history data, concentration history data, time history data, system state history data, patient physiological history data and real-time adjustment error data; Receive and set the initial parameters of the genetic algorithm, including the population size, genetic generation, crossover probability, and mutation probability of the genetic algorithm; An initial population is created, in which each individual represents a potential parameter set of the anesthetic bullet flow regulation model, and the parameter set of the anesthetic bullet flow regulation model is randomly generated; Fitness evaluation, for each individual in the population, each individual is each parameter set, each parameter set is evaluated using the to-be-processed adjustment data set, the evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb, and calculating the error between the model output and the actual regulation demand, and assigning a fitness value to each individual according to the error between the model output and the actual regulation demand; Selection operation, which selects a preset number of individuals as parents from the population through the tournament selection algorithm according to the fitness value; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation: perform mutation operation on offspring individuals to generate new individuals; Merge the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population; Repeat the selection operation, crossover operation and mutation operation until the predetermined genetic generation number is obtained or the minimum error range is met, and then stop the iterative optimization; After the iterative optimization process is completed, the individual with the highest fitness value is selected from the final population. The individual with the highest fitness value is the optimal parameter set, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model.

[0081] The present invention provides a flow regulation system and method for an anesthetic bomb, aiming to solve the problem that the existing flow regulation method of an anesthesia machine cannot accurately adjust the control parameters according to the adjustment requirements. The specific technical scheme and its solution are as follows: Obtain historical data on anesthetic cartridge regulation, including flow, pressure, concentration, time, system status, and patient physiological data, to ensure the comprehensiveness and diversity of the data. Perform feature extraction on the preprocessed data to extract the flow features, pressure features, concentration features, and time features that are most relevant to anesthetic cartridge flow regulation. Use the extracted feature data to train the deep learning model and build a flow regulation model for anesthetic cartridges. This model can learn the complex mapping relationship between features and flow regulation results.

[0082] Receive anesthetic bomb flow regulation tasks, which include specific target flow regulation ratios. Substitute the target flow regulation ratios into the trained anesthetic bomb flow regulation model and output accurate flow regulation parameters. These parameters are calculated based on the mapping relationship learned by the model and can more accurately meet the regulation requirements.

[0083] The flow rate adjustment parameters are sent to the anesthetic bomb adjustment terminal to collect the flow rate, pressure and other data of the anesthetic bomb in real time. These real-time data are compared with the flow rate adjustment parameters to calculate the adjustment error data. This step can provide real-time feedback on the adjustment effect and provide a basis for subsequent model optimization.

[0084] The real-time adjustment error data is integrated with the historical data of the anesthetic bullet adjustment, and the model is optimized using a genetic algorithm. By simulating natural selection and genetic mechanisms, the genetic algorithm can efficiently search and optimize model parameters. Through operations such as selection, crossover, and mutation, the model parameters are iteratively optimized until the predetermined genetic generation is reached or the minimum error range is met. This step can further improve the accuracy and stability of the model.

[0085] The optimized model is used to process the target flow rate adjustment ratio to obtain more accurate flow rate adjustment parameters. These parameters can more accurately control the flow rate of the anesthetic bomb to meet the needs of medical scenarios such as surgery.

[0086] Through these steps, the present invention provides a method based on data-driven machine learning, which can more accurately output the flow adjustment parameters of the anesthetic bomb according to the adjustment requirements, thereby solving the problem that the existing flow adjustment method of the anesthesia machine cannot obtain accurate adjustment control parameters.

Claims

1. A flow regulation system for an anesthetic bomb, characterized in that: include: A data acquisition unit, used to acquire historical data of anesthetic bomb adjustment, the historical data of anesthetic bomb adjustment includes historical data of anesthetic bomb adjustment flow rate, historical data of pressure, historical data of concentration, historical data of time, historical data of system status and historical data of patient physiology; A model building unit is used to preprocess the historical data of anesthetic bullet adjustment to obtain the preprocessed historical data of anesthetic bullet adjustment, extract features from the historical data of anesthetic bullet adjustment, use the feature data of anesthetic bullet adjustment, and train a deep learning model using the feature data of anesthetic bullet adjustment to obtain a flow regulation model of anesthetic bullet; A task receiving unit is used to receive a flow regulation task of an anesthetic bomb, wherein the flow regulation task of the anesthetic bomb includes a target flow regulation ratio, and the target flow regulation ratio is substituted into a flow regulation model of the anesthetic bomb, and the flow regulation model of the anesthetic bomb outputs a flow regulation parameter; The adjustment parameter comparison unit is used to send the flow adjustment parameter to the anesthetic bomb adjustment terminal, collect the real-time data of the anesthetic bomb adjustment terminal, compare the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameter, and obtain real-time adjustment error data; The adjustment parameter optimization unit is used to optimize the flow adjustment model of the anesthetic bullet through a genetic algorithm using the real-time adjustment error data and the historical adjustment data of the anesthetic bullet to obtain the optimized flow adjustment model of the anesthetic bullet, and use the optimized flow adjustment model of the anesthetic bullet to perform data processing on the target flow adjustment ratio to obtain the optimized flow adjustment parameter.

2. The flow regulating system of the anesthetic bomb according to claim 1, characterized in that: The model building unit is further used for: Preprocessing the acquired historical data of anesthetic bullet adjustment, wherein the preprocessing process includes performing duplicate data detection on the historical data of anesthetic bullet adjustment, performing duplicate data removal processing on the duplicate data, and obtaining the preprocessed historical data of anesthetic bullet adjustment; The preprocessed anesthetic bullet adjustment historical data is subjected to feature extraction to obtain anesthetic bullet adjustment feature data, which includes anesthetic bullet adjustment flow feature, anesthetic gas pressure feature, anesthetic gas concentration feature and anesthetic time feature. The extracted anesthetic bullet adjustment feature data is used to train a deep learning model to obtain an anesthetic bullet flow adjustment model.

3. The flow regulating system of the anesthetic bomb according to claim 1, characterized in that: The adjustment parameter comparison unit is further used for: The real-time data of the anesthetic bomb regulating terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration; The flow adjustment parameters include a desired anesthetic gas flow rate, a desired anesthetic gas pressure, and a desired anesthetic gas concentration; Compare the real-time data of the anesthetic bomb regulating terminal with the flow regulating parameters, including the comparison between the real-time anesthetic gas flow and the expected anesthetic gas flow, the comparison between the real-time anesthetic gas pressure and the expected anesthetic gas pressure, and the comparison between the real-time anesthetic gas concentration and the expected anesthetic gas concentration; The difference between the real-time value and the expected value of each indicator in the process of data comparison between the real-time data of the anesthetic bomb adjustment terminal and the flow adjustment parameter is calculated to obtain the adjustment error data. The adjustment error data includes the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

4. The flow regulating system of the anesthetic bomb according to claim 1, characterized in that: The adjustment optimization unit is further used for: Integrate the real-time adjustment error data with the anesthetic bomb adjustment history data to obtain an adjustment data set to be processed, wherein the processed adjustment data set includes anesthetic bomb adjustment flow history data, pressure history data, concentration history data, time history data, system state history data, patient physiological history data and real-time adjustment error data; Receive and set the initial parameters of the genetic algorithm, including the population size, genetic generations, crossover probability, and mutation probability of the genetic algorithm; An initial population is created, in which each individual represents a potential parameter set of the anesthetic bullet flow regulation model, and the parameter set of the anesthetic bullet flow regulation model is randomly generated; Fitness evaluation, for each individual in the population, each individual is each parameter set, each parameter set is evaluated using the to-be-processed adjustment data set, the evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb, and calculating the error between the model output and the actual regulation demand, and assigning a fitness value to each individual according to the error between the model output and the actual regulation demand; Selection operation, which selects a preset number of individuals as parents from the population through the tournament selection algorithm according to their fitness values; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation: perform mutation operation on offspring individuals to generate new individuals; Merge the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population; Repeat the selection operation, crossover operation and mutation operation until the predetermined genetic generation number is obtained or the minimum error range is met, and then stop the iterative optimization; After the iterative optimization process is completed, the individual with the highest fitness value is selected from the final population. The individual with the highest fitness value is the optimal parameter set, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model.

5. A flow rate regulating method for an anesthetic bullet, applied to the flow rate regulating system for an anesthetic bullet as claimed in any one of claims 1 to 4, characterized in that: include: Step S101, obtaining historical data of anesthetic bomb adjustment, the historical data of anesthetic bomb adjustment including historical data of anesthetic bomb adjustment flow rate, historical data of pressure, historical data of concentration, historical data of time, historical data of system status and historical data of patient physiology; Step S102, preprocessing the anesthetic bullet adjustment history data to obtain the preprocessed anesthetic bullet adjustment history data, extracting features from the anesthetic bullet adjustment history data, using the anesthetic bullet adjustment feature data, and training a deep learning model using the anesthetic bullet adjustment feature data to obtain a flow regulation model for the anesthetic bullet; Step S103, receiving an anesthetic bomb flow regulation task, the anesthetic bomb flow regulation task including a target flow regulation ratio, substituting the target flow regulation ratio into a flow regulation model of the anesthetic bomb, and the flow regulation model of the anesthetic bomb outputs a flow regulation parameter; Step S104, sending the flow adjustment parameter to the anesthetic bomb adjustment terminal, collecting real-time data of the anesthetic bomb adjustment terminal, comparing the real-time data of the anesthetic bomb adjustment terminal with the flow adjustment parameter, and obtaining real-time adjustment error data; Step S105, optimizing the flow regulation model of the anesthetic bomb by using the real-time adjustment error data and the historical adjustment data of the anesthetic bomb through a genetic algorithm to obtain an optimized flow regulation model of the anesthetic bomb, and using the optimized flow regulation model of the anesthetic bomb to perform data processing on the target flow regulation ratio to obtain an optimized flow regulation parameter.

6. The flow regulating system of the anesthetic bomb according to claim 1, characterized in that: The step S102 includes: Preprocessing the acquired historical data of anesthetic bullet adjustment, wherein the preprocessing process includes performing duplicate data detection on the historical data of anesthetic bullet adjustment, performing duplicate data removal processing on the duplicate data, and obtaining the preprocessed historical data of anesthetic bullet adjustment; The preprocessed anesthetic bullet adjustment historical data is subjected to feature extraction to obtain anesthetic bullet adjustment feature data, which includes anesthetic bullet adjustment flow feature, anesthetic gas pressure feature, anesthetic gas concentration feature and anesthetic time feature. The extracted anesthetic bullet adjustment feature data is used to train a deep learning model to obtain an anesthetic bullet flow adjustment model.

7. The flow regulating system of the anesthetic bomb according to claim 1, characterized in that: The step S104 includes: The real-time data of the anesthetic bomb regulating terminal includes real-time anesthetic gas flow, real-time anesthetic gas pressure and real-time anesthetic gas concentration; The flow adjustment parameters include a desired anesthetic gas flow rate, a desired anesthetic gas pressure, and a desired anesthetic gas concentration; Compare the real-time data of the anesthetic bomb regulating terminal with the flow regulating parameters, including the comparison between the real-time anesthetic gas flow and the expected anesthetic gas flow, the comparison between the real-time anesthetic gas pressure and the expected anesthetic gas pressure, and the comparison between the real-time anesthetic gas concentration and the expected anesthetic gas concentration; The difference between the real-time value and the expected value of each indicator in the process of data comparison between the real-time data of the anesthetic bomb adjustment terminal and the flow adjustment parameter is calculated to obtain the adjustment error data. The adjustment error data includes the actual value being higher than the expected value, the actual value being lower than the expected value, and the actual value being consistent with the expected value.

8. The flow regulating system of the anesthetic bomb according to claim 1, characterized in that: The step S105 includes: Integrate the real-time adjustment error data with the anesthetic bomb adjustment history data to obtain an adjustment data set to be processed, wherein the processed adjustment data set includes anesthetic bomb adjustment flow history data, pressure history data, concentration history data, time history data, system state history data, patient physiological history data and real-time adjustment error data; Receive and set the initial parameters of the genetic algorithm, including the population size, genetic generations, crossover probability, and mutation probability of the genetic algorithm; An initial population is created, in which each individual represents a potential parameter set of the anesthetic bullet flow regulation model, and the parameter set of the anesthetic bullet flow regulation model is randomly generated; Fitness evaluation, for each individual in the population, each individual is each parameter set, each parameter set is evaluated using the to-be-processed adjustment data set, the evaluation process includes substituting each parameter set into the flow regulation model of the anesthetic bomb, and calculating the error between the model output and the actual regulation demand, and assigning a fitness value to each individual according to the error between the model output and the actual regulation demand; Selection operation, which selects a preset number of individuals as parents from the population through the tournament selection algorithm according to their fitness values; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation: perform mutation operation on offspring individuals to generate new individuals; Merge the offspring individuals generated after crossover and mutation operations with the remaining unselected parent individuals to form a new population; Repeat the selection operation, crossover operation and mutation operation until the predetermined genetic generation number is obtained or the minimum error range is met, and then stop the iterative optimization; After the iterative optimization process is completed, the individual with the highest fitness value is selected from the final population. The individual with the highest fitness value is the optimal parameter set, and the optimal parameter set is used as the parameters of the optimized anesthetic bullet flow regulation model.

Images