论文部分内容阅读
癫痫是大脑神经元异常放电所引起的常见神经系统疾病,其发作具有突然性和反复性特点,因此,提前预测发作以便对患者及时采取措施具有重要意义。本文引入符号动力学方法分析癫痫大鼠失神性发作时脑电(EEG)信号的特性,并对影响符号统计量的关键参数的选取进行讨论,计算癫痫发作不同时期EEG信号的符号熵和时间不可逆转性。研究发现正常发作间隙期,符号熵和时间不可逆转性指标值较大;从发作间隙期向发作期的转化阶段,即发作前期,二者明显减小;发作时维持较低水平。研究结果表明符号动力学方法能够揭示癫痫EEG动力学特征变化,符号熵和时间不可逆转性两个指标是表征癫痫发作不同阶段的敏感特征量,具有重要的潜在临床应用价值。
Epilepsy is a common neurological disease caused by abnormal discharge of cerebral neurons. The onset of epilepsy is characterized by suddenness and repetitiveness. Therefore, it is of great significance to predict the onset in advance so that the patients can take timely measures. This paper introduces the symbolic dynamics method to analyze the characteristics of EEG signals in epileptic rats during absence of dexterity and discusses the selection of the key parameters that affect the sign statistics. The calculation of symbol entropy and time of EEG signals in different periods of epileptic seizure can not be made Reversal. The study found that the gap between the normal seizure, symbolic entropy and time irreversibility index larger; from the onset of seizure period to the stage of the transition period, that is, the early onset, both significantly reduced; seizure to maintain a low level. The results show that the symbolic dynamics method can reveal the dynamic characteristics of epilepsy EEG, and the two indexes of symbolic entropy and time irreversibility are sensitive characteristics of different stages of epileptic seizures, which has important potential clinical value.