利用广义复杂度对癫痫病发作进行预测的方法研究

摘要

This article aims at predicting the epileptic seizures through observing the changes of EEG (electroencephalogram) signal' s complexity of epileptic patients, by means of generalized complexity. Methods: We use the digital signals to calculate the complexity of every interval by means of generalized complexity computation. By observing and calculating the complexity of EEG data, we can get the changing state of complexity of patients' EEG data before the epileptic seizure. Results: Through the calculation of complexity of 21 epileptic patients' EEG digital signal in the period of seizure, we observed the changing state of the complexity of these signals, and realized the prediction of epileptic seizure. The average accuracy reached 94.3% through the prediction research of the data of 21 epileptic patients. Conclusions: This article proposes a means of predicting the spasm of epilepsy based on the generalized complexity. The experiment result shows that the method proposed in this article can foresee the spasm of epilepsy.%目的:通过用广义复杂度的运算方法,分析癫痫病患者发病期间的脑电图,计算出这一段脑电图数据的复杂度,通过对复杂度数据的分析,对癫痫病的发作进行预测.进而提高癫痫病患者的生活质量,降低癫痫病患者的生命危险.方法:将癫痫病发病的脑电图的图像信号还原成数字信号,计算出该数字信号各个区间的广义复杂度值.得出癫痫病发病前脑电图数据复杂度的变化状态,对癫痫病患者发病进行预测.结果:计算出21名癫痫病患者在发病前后共一个小时的脑电图数字信号的广义复杂度值,通过观察21名癫痫病患者在癫痫病发病期间的脑电图的数字信号的广义复杂度变化情况,对癫痫病的发作进行预测(对21名病人癫痫病发作的数据进行预测研究,平均准确率可达94.3％).结论:本文提出了一种基于广义复杂度的癫痫发作预测方法,通过对21名癫痫病患者的脑电图信号的实验和计算,结果表明,该方法可以有效地预测到癫痫发作.