FPGA Implementation of Single Neuron PID Controller for Depth of Anesthesia Based on PSO

• Published in: 2018 Third Scientific Conference of Electrical Engineering (SCEE) pp. 247 - 252
• Main Authors: Abood, Layla H., Karam, Ekhlas H., Issa, Abbas H.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2018
• Subjects: FPGA; Neural Network; Particle Swarm Optimization; PID controller
• DOI: 10.1109/SCEE.2018.8684186

Anesthesia is considered as one of the most important part of any surgical operation due to this it must be good monitored and controlled, i.e. the rate of infusion must be given an inappropriate dose to save the patient status in an adequate level of anesthesia. The model translates the relation between the drug used and patient response is pharmacokinetic pharmacodynamic (PK/PD). Bispectral index (BIS) is the most standard for maintaining the Anesthesia level. In this paper an optimal PID controller with neural network consists of one neuron is presented to evaluate the drug infusion rate, the system input will be the propofol drug rate, and the level of hypnosis will be the system output since it is measured during surgery from the monitor of the BIS device. optimal controller will calculate the infusion rate by using particle swarm optimization (PSO) method for tuning all the variables needed to find the optimal rate and considered it as a control signal. Then all the result is obtained by Matlab software and the hardware implantation is done using the Integrated Synthesis Environment from Xilinx ISE version 14.6 and downloaded on Spartan 3 Kit (XC3SD3400ACS484-4 by). The output results obtained from the controller in measuring the value of the infusion rate and tracking the BIS value for different patients show its efficiency and robustness in measuring the Depth of Anesthesia (DOA) correctly.