基于Pinsky-Rinzel模型癫痫状态的估计与闭环混合控制  

Estimation and Closed-Loop Mixed Control of Epileptic State Based on Pinsky-Rinzel Model

作　　者：曹玉珍[1] 张丽媛[1] 刘晨[2] 王江[2] 

机构地区：[1]天津大学精密仪器与光电子工程学院,天津300072 [2]天津大学电气与自动化工程学院,天津300072

出　　处：《纳米技术与精密工程》2015年第3期232-238,共7页Nanotechnology and Precision Engineering

基　　金：国家自然科学基金资助项目(61374182);天津市自然科学基金重点资助项目(13JCYBJC17600)

摘　　要：癫痫是一种大脑神经元异常同步放电的神经系统疾病,其放电与神经细胞外离子浓度及神经元内部耦合强度等关键参数有关.本文通过无迹卡尔曼滤波器(UKF)实现Pinsky-Rinzel(PR)模型的关键参数估计,提出一种延时反馈与比例-积分(PI)控制相结合的混合控制,实现PR模型的放电模式转换,消除神经元癫痫放电状态.采用数值仿真进行分析,证明所提出的混合控制方法能够输出较小的刺激电流,快速实现癫痫状态的转换.Epilepsy is a neurological disease caused by the abnormal synchronous firing of brain neurons. The morbid firing state is related to such critical parameters as the ionic concentration outside neurons and the coupling strength inside neurons. Electrical stimulation is an effective therapy in the treatment of epi- lepsy. This paper adopts unscented Kalman filter （UKF） to estimate the critical parameters of Pinsky- Rinzel （PR） model. Then a closed-loop mixed control system consisting of delayed feedback and propor- tion-integration （PI） control is proposed to automatically switch the epileptic firing pattern, suppressing the neurons from firing. By analyzing the numerical simulation results, the strategy proposed could pro- vide a stable stimulating current more timely, and suppress the epilepsy seizure more effectively.

关 键 词：闭环控制 无迹卡尔曼滤波器 电刺激 Pinsky—Rinzel模型 癫痫状态 

分 类 号：R742.1[医药卫生—神经病学与精神病学] TN911.7[医药卫生—临床医学]