论文部分内容阅读
对麻醉的SD大鼠在癫痫发作前后两种状态的皮层脑电 (ECoG)的时间序列 ,用多种有效的方法和分析技术 ,使得大量的ECoG时间序列得以正确的分析 ,并得出重要的结论 .首先利用延时坐标法重构吸引子 ;计算互信息函数 ,取互信息函数第一次达到最小值的延时为重构延时时间 ,提出将伪邻点法和Cao法相结合的方法确定最佳嵌入维数 .然后采用非线性预报和替代数据法相结合的方法确定ECoG为混沌时间序列 ,从不同角度得出了ECoG不是低维混沌的结论 .在ECoG相空间重构的基础上 ,计算了最大Lyapunov指数 (LLE) .应用了近似熵这一标量对ECoG进行刻画 ,计算结果表明 :癫痫发作前的皮层脑电的最大Lyapunov指数和近似熵都明显地高于癫痫发作后的 ,这可能为理解癫痫发病机理 ,预报癫痫发作和治疗提供一定的思路 .
Time-series analysis of cortical electroencephalography (ECoG) in both states before and after epileptic seizures in anesthetized SD rats, using a variety of effective methods and analytical techniques, enabled a large number of ECoG time series to be correctly analyzed and concluded that important Conclusions Firstly, the attractor is reconstructed by using time-delay coordinate method. The mutual information function is calculated, and the delay for the first time when the mutual information function reaches the minimum value is taken as the reconstruction delay time. A method combining the pseudo-neighbor method with the Cao method Then determine the optimal embedding dimension.And then use the combination of nonlinear forecasting and alternative data to determine ECoG as chaotic time series and draw conclusions from different angles that ECoG is not low dimensional chaos.On the basis of ECoG phase space reconstruction, The maximum Lyapunov exponent (LLE) was calculated.The ECoG was described by using the approximate entropy, and the results showed that the maximum Lyapunov exponent and approximate entropy of cortical EEG before epileptic seizure were significantly higher than those after seizure It may provide some ideas for understanding the pathogenesis of epilepsy, forecasting seizures and treatment.