基于EEG高阶过零分析的脑损伤检测新方法

HIGHER ORDER CROSSING ANALYSIS OF EEG FOR BRAIN INJURY QUANTIFICATION

目的 :提出基于高阶过零 (HOC)的方法并利用EEG检测由于缺氧窒息而引起的中枢神经系统损伤及对损伤程度进行定量评估。方法 :采用两种基于HOC的方法 ,一种为依据EEG信号序列统计分布特性的HOCSD(S距离 )法 ,另一种为依据预处理后EEG信号的谱特性的HOCCF法。结果 :对三类实验样本进行了检测 ,检测结果与神经缺欠值 (NDS)的相关性高达 96 %。结论 :实验数据分析表明 ,两种基于HOC的新方法对于检测由于缺氧窒息而引起的中枢神经系统损伤及对损伤程度进行定量评估具有一定的意义。

Objectives: To develop two higher order crossings (HOC) based EEG analysis methods for the brain injury detection and quantification. Methods: Two HOC based methods were used. One was based on the statistical distribution of the EEG time series via a distance measure. The other was based on the spectrum characteristics of a preprocessed EEG signal. Results: The detection and analysis results of three types of samples showed that the correlation coefficients between the results and the neurological deficit scores (NDS) obtained from several behavioral tests were as high as 96%. Conclusions: The new HOC-based methods are more correlated with the NDS than existing methods. They are useful for the brain injury detection and analysis of the central nervous system.