Optimization Of Multi-channel Transcranial Magnetic Stimulation Coil Array Based On Improved Particle Swarm Algorithm

Transcranial magnetic stimulation(TMS)is a novel treatment method for mental illness with non-invasive,painless,showing great potential for development in biomedical and rehabilitation medicine.Multichannel transcranial magnetic stimulation is a flexible system of TMS.Its stimulation source is a coil array composed of multiple magnetic stimulation coils.It has the advantages of simultaneous multi-target stimulation and strong controllability.However,how to reasonably control the excitation current of the coil array and realize the precise focusing of the induced electric field at the target point has always been one of the difficulties preventing the multichannel transcranial magnetic stimulation from being widely used.Aiming at the problem that multichannel transcranial magnetic stimulation cannot accurately achieve the precise focus of the induced electric field at the target point,a novel dynamic multi-group particle swarm optimization algorithm NMSPSO is proposed based on the dynamic multi-group particle swarm optimization algorithm,which has three improved strategies: novel information exchange strategy,classification learning strategy and mutation strategy.The effectiveness of the three improved strategies is verified by simulation,the effect of parameter settings on the performance of the algorithm is analyzed,and comparison with several other particle swarm algorithm variants.The results indicate that NMSPSO algorithm has better solution accuracy and faster convergence speed.Experiments on the embedded Arduino platform prove that the NMSPSO algorithm has higher solution efficiency while ensuring the accuracy of the solution.To verify the effect of the improved algorithm on the electric field focusing optimization of multichannel coil array,a joint simulation platform is established by COMSOL and MATLAB,which realizes the joint simulation of physical field and intelligent optimization algorithm.The double layer coil array model and the five layer head model are established,and the mesh is divided.NMSPSO algorithm and several other particle swarm optimization algorithms are used to optimize the electric field focusing of the double layer coil array based on the joint simulation platform.The results show that the electric field optimized by NMSPSO algorithm can better focus around the target,and the focus is improved by 6%～24% compared with several other algorithms.It is proved that the focus of the induced electric field at the target can be improved by using the algorithm proposed in this dissertation to optimize the multichannel coil array.This dissertation provides a novel optimization strategy for improving the electric field focus of the multichannel transcranial magnetic stimulation system at the target point,which has a guiding effect on the further development of multichannel transcranial magnetic stimulation.