Particle swarm optimization algorithm based parameters estimation and control of epileptiform spikes in a neural mass model

Design of Closed-Loop Control Schemes Based on the GA-PID and GA-RBF-PID Algorithms for Brain Dynamic Modulation

Neurostimulation can be used to modulate brain dynamics of patients with neuropsychiatric disorders to make abnormal neural oscillations restore to normal. The control schemes proposed on the bases of neural computational models can predict the mechanism of neural oscillations induced by neurostimulation, and then make clinical decisions that are suitable for the patient's condition to ensure better treatment outcomes. The present work proposes two closed-loop control schemes based on the improved incremental proportional integral derivative (PID) algorithms to modulate brain dynamics simulated by Wendling-type coupled neural mass models. The introduction of the genetic algorithm (GA) in traditional incremental PID algorithm aims to overcome the disadvantage that the selection of control parameters depends on the designer's experience, so as to ensure control accuracy. The introduction of the radial basis function (RBF) neural network aims to improve the dynamic performance and stability of the control scheme by adaptively adjusting control parameters. The simulation results show the high accuracy of the closed-loop control schemes based on GA-PID and GA-RBF-PID algorithms for modulation of brain dynamics, and also confirm the superiority of the scheme based on the GA-RBF-PID algorithm in terms of the dynamic performance and stability. This research of making hypotheses and predictions according to model data is expected to improve and perfect the equipment of early intervention and rehabilitation treatment for neuropsychiatric disorders in the biomedical engineering field.

Closed-loop seizure modulation via extreme learning machine based extended state observer

Neuromodulation is a promising way in clinical treatment of epilepsy, but the existing methods cannot adjust stimulations according to patients' real-time reactions. Therefore, it is necessary to acquire a systematic and a scientific regulation method based on patients' real-time reactions. The linear active disturbance rejection control can adapt to complex epileptic dynamics and improve the epilepsy regulation, even if little model information is available, and various uncertainties and external disturbances exist. However, a linear extended state observer estimates the time-varying total disturbance with a steady-state error. To improve regulation, it is crucial to estimate the total disturbance in a more accurate manner. An extreme learning machine is capable of approximating any nonlinear function. Its initial parameter generation is more convenient, adjustable parameters are fewer, and learning speed is faster. Thus, a nonlinear time-varying function can be estimated more timely and accurately. Then, an extreme learning machine based extended state observer is proposed to get a more satisfactory total disturbance estimation and more desired closed-loop regulation. The convergence of the extreme learning machine based extended state observer is verified and the stability of the closed-loop system is analyzed. Numerical results show that the proposed extended state observer is much better than a linear extended state observer in estimating the total disturbance. It guarantees a more satisfied closed-loop neuromodulation.