改进的相对转移熵的癫痫脑电分析  

Analysis on relative transfer of entropy based on improved epileptic EEG

作　　者：王莹[1] 侯凤贞[2] 戴加飞[3] 刘新峰[3] 李锦[4] 王俊[1] 

机构地区：[1]南京邮电大学,图像处理与图像通信江苏省重点实验室,南京210003 [2]中国药科大学理学院,南京210009 [3]南京军区南京总医院神经内科,南京210002 [4]陕西师范大学物理学与信息技术学院,西安710062

出　　处：《物理学报》2014年第21期413-419,共7页Acta Physica Sinica

基　　金：国家自然科学基金(批准号:61271082;61201029;61102094);江苏省自然科学基金(批准号:BK2011759;BK2011565)资助的课题~~

摘　　要：脑电信号是由脑神经活动产生并且始终存在于中枢神经系统的自发性电位活动,是一种重要的生物电信号.脑电信号是非常微弱的且是非线性的,脑电信号也具有时间不可逆性.本文提出了一种新的基于正向序列转移概率与逆向序列转移概率的相对熵方法即相对转移熵方法,并应用此方法研究了正常脑电与癫痫脑电的不可逆性,实验结果显示癫痫患者的脑电信号的不可逆性明显小于正常人的脑电信号的不可逆性.这说明改进的相对转移熵可以作为一个物理过程不可逆程度的度量参数,这使得应用脑电信号区分病人是否患有癫痫疾病具有积极指导意义.EEG（electroencephalogram） is generated by the brain activity and is present in the central nervous system of spontaneous electrical activity, which is an important biological signal. EEG is a very weak and nonlinear as well as irreversible signal. This paper presents a new method to describe it based on the relative entropy of transition probability for the forward and reverse sequences. Besides, we may apply this method to study the normal EEG and epileptic EEG irreversibility, and the experimental results show that the EEG irreversibility of patients who suffer from epilepsy is significantly less than that of normal people. This shows that the relative transfer entropy can be used as aparameter to detect the irreversible degree of EEG for recognizing whether a patient is suffering from epilepsy or not, which may be a positive index for clinical diagnosis.

关 键 词：相对转移熵 脑电信号 符号化 时间不可逆性 

分 类 号：R742.1[医药卫生—神经病学与精神病学]