癫痫大鼠脑电信号的除趋势涨落分析

Detrended Fluctuation Analysis of Epileptic Rat EEG

除趋势涨落分析(Detrended fluctuation analysis,DFA)是一种较新的非稳态时间序列长时程暂态相关(Long-range temporal correlation,LRTC)结构特征分析方法,人们已将它初步应用于正常脑电信号和疾病脑电信号的相关性分析之中,取得了一些有意义的结论。我们用DFA方法分析了匹罗卡品痫样放电大鼠模型的不同脑区和不同脑功能状态下脑电演化行为的参数——标度指数,结果表明:癫痫大鼠从无痫样放电到连续痫样放电再到周期痫样放电的过程转变中,4个脑区(左皮层、右皮层,左海马、右海马)的标度因子分别从约0.97到0.82再到0.94变化,配对样本T检验表明,同一脑区不同脑功能状态间的标度因子有显著性差异(P<0.05),提示这三种脑功能状态的局部神经网络动力学可能不同。对三种不同的脑功能状态,标度因子在不同脑区上没有显著性差异(P>0.05)。

Detrended fluctuation analysis （DFA） is a new technique used to characterize the long-range temporal correlation （ LRTC） structure of non-stationary time series, which has been applied in electroencephalographic （ EEG） oscillations and diseases such as Alzheimer and stroke. In this paper, DFA is used to study the scaling exponents of intracranial EEG recordings of Pilocarpine-induced epileptic rat. It was found that when the brain functional status changed from nonepileptiform discharges to continuous-epileptiform discharges and to period-epileptiform discharges, the average scaling exponents of four brain regions （left cortex and left hippocarnpus, right cortex and right hippocarnpus） changed from 0,97 to 0.82 and to 0.94, and a statistically significant difference between the brain functional states was shown by paired T test （ P〈 0,05 ）, thus suggesting that the local neuronal network dynamics of the three brain function states may be different. But in regard to the three brain function states, there is no statistically significant difference between the four brain regions（ P 〉0.05 ）.