The potential application of entropy measurements to analyze short-term Electrocephalograph (EEG) traces of patients with temporal lobe epilepsy (TLE) was investigated.Total eight channels EEG signals, collected from 8 patients and 10 healthy subjects, were analyzed by both algorithms of the approximate entropy (ApEn) and sample entropy (SampEn).Two sliding windows and 5 different filters levels r were used and discussed. The entropies of EEG were significantly lower in patients with TLE than that in healthy one. The degree of complexity in the epileptic focus hemisphere was lower than in the non-focus hemisphere in patients. Small sliding window may provide more details associated with the seizure. The filter level r must not be smaller than 0.15% SD for 1 s-window whilst it must not be smaller than 10% SD for 4 s-window. The results have demonstrated that entropy measurements could be alternative nonlinear approaches for analyzing short-term EEG signals. The observed lower values in the complexity of the EEG signal for patients with TLE provide preliminary support for the notion that the complex nonlinear nature of brain electrical activity may be the result of isolation or impairment of the neural information transmission within the brain.%采用了近似嫡(tapproximtely entropy,ApEn)和它的改进算法,即样品熵(sample entropy,SampEn)分析了8位颞叶癫痫患者和10位健康人员的短程脑电信号.在计算过程中使用了两种滑动窗口和5个不同的过滤标准r.结果显示小的滑动窗口能更多地反映与阗痫发作相关的细节.对于1秒的滑动窗口,过滤标准r不能小于时间序列标准差的0.15%;而对于4秒的滑动窗口,则过滤标准r不能小于时间序列标准差的10%.研究结果表明,在短程电信号的非线性分析中,样品熵是一种比近似熵更为可靠的非线性分析方法.颞叶癫痫患者的脑电信号的熵值低于健康人员,这可能表明脑电活动的非线性程度的隆低是由于神经信号在大脑内的传递受到了阻碍或者损坏,使得神经信号成为孤立的信息源.
展开▼
