We will study the mesoscopic dynamics underlying mesiobasal temporal lob epilepsy (MBTLE) and compare the results with similar analyses of individuals with normal (specifically non-epileptic) brain function. Preliminary results indicate the following: At the level of 10**5 or more neurons, and over a range of time horizons, a model independent analysis indicates that the intracranial and scalp recordings from patients with MBTLE are nonstationary, and that the nonstationary can be explained by the time dependence of a single parameter. Moreover, the brains of MBTLE patients can be characterized by fluctuations between two dynamical states. Transitions between these two states are observable in both intracranial and scalp electrodes during both interictal periods (more than two hours away from a clinical seizure) and preictal periods (within 30 minutes preceding a seizure). The time dependent parameter indicated by the model independent non-stationarity analysis can be interpreted as a time-dependent transition probability between these two states. As a seizure approaches the frequency and nature of the transition probabilities changes, the changes being observable in scalp recordings at least 30 minutes prior to the seizure. Analyses of preliminary data from scalp recordings of normals (non- epileptics) show no evidence of the non-stationarity observed in epileptics. In addition, the normal data is well described by a single dynamical state. Thus, the epileptic brains appear to be bistable with a time dependent transition probability while normal brains appear to be describable by a single state. In addition, there appears to be uniformity of the nonstationarity across electrodes, suggesting that the bistability observed in single leads in MBTLE patients may be a signature of complex spatio-temporal excitations that traverse the brain during interictal and preictal periods. It is the purpose of this proposal to verify and extend these observations, specifically with an eye toward understanding the mesoscopic dynamics of the epileptic brain and how they differ from the normal brain. In addition, the observation of systematic changes in the nonstationarity as a seizure approaches, even in scalp recordings, suggests possible strategies for predicting seizure onset early enough for interventive or other actions to reduce the dangers and effects of the seizure.
| Li, Dingzhou; Zhou, Weiping; Drury, Ivo et al. (2006) Seizure anticipation, states of consciousness and marginal predictability in temporal lobe epilepsy. Epilepsy Res 68:9-18 |
| Li, Dingzhou; Zhou, Weiping; Drury, Ivo et al. (2003) Non-linear, non-invasive method for seizure anticipation in focal epilepsy. Math Biosci 186:63-77 |
| Li, Dingzhou; Zhou, Weiping; Drury, Ivo et al. (2003) Linear and nonlinear measures and seizure anticipation in temporal lobe epilepsy. J Comput Neurosci 15:335-45 |
| Savit, R; Li, D; Zhou And, W et al. (2001) Understanding dynamic state changes in temporal lobe epilepsy. J Clin Neurophysiol 18:246-58 |
