We propose to investigate the dynamical properties of the epileptogenic focus in patients with temporal lobe epilepsy. Recently developed techniques used in the study of complex nonlinear systems will be employed to analyze the electroencephalographic signal generated by the epileptogenic focus. These techniques are able to distinguish and quantitatively characterize important properties of potential predictive and clinical significance of brain electrical activity not possible by expert visual analysis or more traditional quantitative measures of the original EEG signal. Our preliminary studies, performed in a limited number of cases, indicate that the dynamical properties of brain electrical activity generated by a seizure focus differ qualitatively and quantitatively among the interictal, ictal and postictal states. These studies also suggest that there is a gradual evolution from the interictal to the ictal state. We wish to verify these findings in a larger number of patients and to apply newer techniques of dynamical analysis that provide a quantitative measure of the predictability of the system over time as well as a more detailed characterization of the dynamics. In our initial studies, electrocorticographic recordings were analyzed. In the proposed study, we will analyze recordings from depth electrodes implanted into the epileptogenic hippocampus. Specifically, we wish to test three hypotheses: (l) The dynamical behavior of the EEG generated by a seizure focus is more complex, less predictable and more chaotic in the interictal state than in the ictal state and is most complex, least predictable and most chaotic in the immediate postictal state; (2) there is a characteristic evolution from the preictal state to the ictal state that can be detected by quantitative dynamical analysis of the EEG signal; and (3) during the interictal state, the dynamical properties of the EEG generated by the seizure focus are quantitatively different than the EEG generated by the homologous region of the contralateral hemisphere. It is anticipated that this research will further our understanding of the physiological disturbances that occur in human epilepsy. It is expected that insights derived from the proposed study will lead to the development of computer- assisted procedures for detecting and localizing the onset of seizures and determining the pathways of propagation to other areas of the brain. Such procedures would have clinical utility in the presurgical evaluation of patients with medically intractable temporal lobe epilepsy.
| Chaovalitwongse, W; Iasemidis, L D; Pardalos, P M et al. (2005) Performance of a seizure warning algorithm based on the dynamics of intracranial EEG. Epilepsy Res 64:93-113 |
| Manuca, R; Casdagli, M C; Savit, R S (1998) Nonstationarity in epileptic EEG and implications for neural dynamics. Math Biosci 147:1-22 |
| Casdagli, M C; Iasemidis, L D; Savit, R S et al. (1997) Non-linearity in invasive EEG recordings from patients with temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol 102:98-105 |
