摘要
针对癫痫脑电信号中识别高频振荡节律困难的问题,提出一种基于Teager能量算子和曲线长度的多导联高频振荡节律(high-frequency oscillations,HFOs)特征提取与病灶定位算法。首先,采用陷波器和FIR数字滤波器对癫痫高频振荡信号进行预处理,其次,利用Teager能量算子和曲线长度对高频振荡节律的特征进行提取。最后,以功率谱为手段进行癫痫病灶定位。实际癫痫高频振荡仿真实验结果验证了该方法的有效性。
Considering that it is difficult to identify the high frequency oscillation issues in epileptic electroencephalogram（EEG）, a multi-lead high frequency oscillation feature extraction and lesion localization algorithm was proposed based on Teager energy operator and curve length. Firstly, a notch filter and FIR digital filters were adopted to preprocess the high frequency oscillation of epilepsy. Secondly, teager energy operator and curve length were used to extract features in high-frequency oscillation. Finally, power spectrum was taken as a means of initial positioning for epileptic foci. The experiment results of actual epilepsy high frequency oscillation data simulation verify the feasibility of the method.
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2015年第9期3302-3308,共7页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(61573381)~~
关键词
癫痫脑电
高频振荡
TEAGER能量算子
特征提取
癫痫病灶定位
eileptic electroencephalogram
high frequency oscillation
Teager energy operator
feature extraction
epilepsy positioning
