| Epilepsy is a neurological disorder characterized by repeated,unprovoked that can cause serious health problems.So far it affects more than 65 million people worldwide,one third of whom cannot be effectively treated through current methods.To address above challenges,the present investigation has sought to develop a novel prediction strategy for seizure detection based on the graph properties of the cortical network.We will discuss the effectiveness of the prediction method from the following two aspects:(1)Nonlinear partially directed coherence(NPDC)was employed as measure of functional brain networks(FBNs)and analyzed to reveal the directional flow of epilepsy-linked brain activity.A novel prediction strategy was then developed for the prediction of epileptic seizures by introducing extracted network features to an extreme learning machine(ELM).Results show that the proposed method achieved favorable performance across all subjects and in all EEG frequency bands,with best accuracy of 89.21% in beta band and an optimal prediction time of 1356.48 seconds in delta bands,which outperforms currently available approaches.(2)A neural mass model(NMM)was employed to simulate brain depth EEG signals and clarify the relationship of network structure,node dynamics,and the generation of epileptic discharge.To quantify the ability of the epileptogenic zone(EZ)to generate emergent pathological dynamics,the index of brain network ictogenicity(BNI)has been introduced.To our knowledge,this represents the first time that BNI has been used as a predictor for epileptic seizures.The results demonstrate that the proposed method achieves favorable performance across the introduced subjects based on Depth EEG data.On average,seizures were predicted 2461.74 seconds before onset,which outperforms many currently available approaches.The findings of this study demonstrate that the proposed prediction strategy is suitable,and that the FBNs is a valuable index of the prediction of seizure onset. |
PDF Full Text Request