Epileptic Eeg Analysis

Epilepsy is a serious disease for human health. Excitability and excessive firing of neurons are improved at seizures, which resulted in transient abnormality of centre nerve system. Epilepsy is clinically diagnosed according to electroencephalogram (EEG) of patient, and evaluation of the side effects of antiepileptic medicine and sequelae of operation must refer to it. In brief, Analysis of EEG is very important during the clinical diagnose, treatment and valuation of epilepsy. EEG is collective performance of neuronal population on the cortex or scalp. During epileptic seizures, EEGs shift from one chaotic state to another. It would provide theoretic evidences for prediction and treatment of epilepsy as well as explanation of epileptic mechanism by nonlinear analysis of EEG in different cerebral function. In our study, clinical epileptic EEGs were analyzed by approximate entropy (ApEn), it was demonstrated that epileptifrom discharges could be detected from background EEG by ApEn. The animal model of rat's Epileptiform discharges was successfully constructed. The ECoGs and EHGs of three kinds of different function states—non-epileptiform discharges, continuous-epileptiform discharges and period-epileptiform discharges were analyzed. Some results were listed as following: 1. Approximate entropy analysis (1) With the approximate entropy analysis of ECoGs and EHGs with frequency of 0.5~30Hz, it was found that when the state transited from non-epileptiform discharges to continuous-epileptiform discharges, ApEn of signal in every brain region was significantly decreased (p<0.01). These results confirm that neuronal action of relative regions trends synchronization at epileptiform discharges.Temporal ApEn analysis demonstrates that epileptiform discharges could be distinguished from background EEG by ApEn. (2) Approximate entropy Analysis of divided rhythm of ECoG and EHG (delta(0.4~4.5Hz), theta(4.5~8Hz), alpha(8~12Hz), sigma(12~16Hz) and beta(16~30Hz)) was analyzed when different states shifted, it was found that when the state transited from non-epileptiform discharges to continuous-epileptiform discharges, ApEn of delta rhythm from cortex was significantly increased (p<0.01), and its relative content of fast wave(2~4.5Hz) component was remarkably added (p<0.05), but that of alpha, sigma and beta rhythm was notably decreased (p<0.05). It indicates that change of complexity could mainly depend on that of high rhythms at transition of cerebral function. If the state transited from continuous-epileptiform discharges to period-epileptiform discharges, ApEn of delta rhythm except was significantly decreased (p<0.01). It suggests that termination of epileptiform discharges might be related to synchronization of low rhythm. 2.Synchronization likelihood Analysis (1)With the synchronization likelihood analysis of ECoG and EHG with frequency of 0.5~30Hz, it was found that when the state transited from non-epileptiform discharges to continuous-epileptiform discharges, synchronization between left-ECoG and left-EHG was significantly strengthened, and similar change had taken place between right-ECoG and right-EHG, left-ECoG and right-ECoG and left-EHG and right-EHG (p<0.05). It clarifies that epileptiform discharges comes from excessive firing of synchronized neuron population. If the state transited from continuous-epileptiform discharges to period-epileptiform discharges, the synchronization was significantly weakened between left-ECoG and left-EHG, and similar change between left-EHG and right-EHG (p<0.01). These results indicate that left hippocampus might play on important role during epileptiform discharges in this animal model. Synchronization likelihood analysis of time course indicates that synchronization likelihood could describe in-phase process of neuronal population among cerebal regions during epileptiform discharges. (2)Synchronization likelihood Analysis of divided rhythm of ECoG and EHG in the same functional state, it was found that for non-epileptiform discharges state, synchronization of delta rhythm was significantly higher than that of alpha, sigma and beta rhythm (p<0.01). It implies that synchronization phenomenon of delta rhythm ,which was similar with epileptiform discharges, comes into being before the others ; for continuous-epileptiform discharges, synchronizations of delta rhythm were significantly lower than that of theta and alpha rhythm (p<0.05). These results imply that synchronizations of relatively high rhythms are primary and synchronized diversity of between delta and the others rhythms could distinguish functional situations. (3) Synchronization likelihood of divided rhythm of ECoG and EHG was analyzed when different states shifted, it was found that when the state transited...