Simulation Study On Closed-Loop Control Of Propofol Anesthesia In Different Age Groups

Anesthesia closed loop control is a very important part of surgical intelligence.At present,there are some applications in adults,but because of the difference of metabolism between the elderly and children,the research on anesthesia control is still challenging.In view of this situation,the control theory is applied to clinical anesthesia,and the pharmacokinetic models of children,adults and the elderly are established to describe the metabolic process of drugs in the body.Finally,the specific control methods of propofol were analyzed for different age groups.The work is as follows:Firstly,considering the difference of metabolism of different ages,the model of different population was established.Based on the adults’ pharmacokinetics pharmacodynamics model,the model of children was established by allometric scaling,and the model of the elderly was obtained by adding more covariates.Based on three models,the considerable controllability of the models was analyzed from the perspective of traditional control.For the closed-loop control system,the Renyi permutation entropy of the two observed features extracted from EEG was selected as the feedback index,and the commonly used parameter identification method of the observed features was introduced.Secondly,based on the collected adults’ clinical data,the specific pharmacodynamic parameters were obtained by two identification methods.On this basis,two control methods,the fuzzy adaptive PID and ant colony optimization PID,were selected from traditional control perspective,and the reinforcement learning used selected based on the modern control theory to control the anesthesia closed loop of the adults model.In addition,artificial disturbance was added in order to test the anti-jamming performance of the system,the tracking performance of the three control strategies were evaluated at the rise time,adjustment time and overshoot,and the performance of the controllers were compared and analyzed from the three aspects of deviation,accuracy and fluctuation degree.The results showed that the ant colony optimization PID controller is better than the other two controllers in terms of accurate performance and stability performance,while the Q learning control strategy has faster response speed,shorter arrival stability time and stronger anti-interference performance than the other two controllers.Then,based on the analysis of the closed-loop control results of adults,the closed-loop control of children adopted ant colony optimization PID and Q learning two control methods.According to the analysis of the simulation results,all the indexes were within the allowable range of clinical practice,can reach the stable range quickly and stably,had acceptable overshoot,and had good performance under the condition of adding disturbance.It showed that these two control strategies can meet the clinical requirements in children anesthesia control.Finally,based on the clinical data of the elderly,ant colony optimization PID and Q learning were also used to control the elderly model.The results showed that the designed ant colony optimization PID controller has good accuracy and stability,but poor response speed,while Q learning control strategy has faster response speed and better anti-interference performance.