基于能量算法的癫痫发作脑电信号分析

摘要

目的 寻找一种脑电分析方法使其能够检测癫痫发作并为脑电定量分析提供理论依据.方法 基于脑电信号的能量算法计算癫痫发作周期中不同阶段的能量,比较发作间期、发作前期、发作期的能量均值变化.结果 用该方法评价了7位病人的22次癫痫发作共135 h脑电记录,其中18次发作的发作期能量较发作间期、发作前期有明显差异.结论 能量算法可以描述发作过程中能量均值的变化趋势,为癫痫发作分析提供一个方法思路和定量分析手段.%Objective To find a way to detect epileptic seizures and provide a theoretical basis for EEG quantitative analysis. Methods The energy of seizures in different stages of the seizure cycle was calculated based on energy algorithm. The mean energy changes of interictal, pre - ictal and ictal were compared. Results The method was evaluated using data from 7 patients, including 135 h EEG records and 22 seizures. The ictal energy was obviously different from interictal and pre - ictal energy in 18 of 22 seizures. Conclusion Energy algorithm can describe the process of mean energy trends during the seizure cycle and provide a method of thinking and a quantitative analysis tool for seizure analysis.