Abstracts

Rationale: About a third of epilepsy patients are medically intractable and are forced to seek other alternatives. Current electrical stimulation therapies have shown to have curative effects in suppressing seizures and reducing overall seizure frequency. However, the optimization for stimulation parameters to maximize therapeutic effects remains a difficult and time-consuming challenge. There is a need for a quantitative method that is able to track intrinsic network dynamics of the epileptogenic network, its evolution towards seizure, and the effects of electrical stimulation on the network. This would serve as a valuable indicator to help optimize stimulation parameters and measure the therapeutic effects of electrical stimulation. Here we propose a network analysis framework to classify brain states involved in epileptogenesis and study the influence of electrical stimulation on the network.

Methods: Adult male and female Wistar rats were used for all experiments. 4-aminopyridine (4AP) was injected into the CA3 area of the right Hippocampus, and local field potentials (LFP) were recorded using a multi-channel probe implanted into the CA1 area. The proposed algorithm creates linear models from the LFPs using a sliding window every 500msec. The linear models are stacked and analyzed to evaluate the nodal influence within the network. The resulting network feature, is then clustered into i) baseline, ii) preictal, iii) ictal, and iv) postictal activity, and projected into a reduced dimensional space. In this network subspace, it is possible to visualize ictal/seizure zones and baseline/healthy zones. A multivariate-gaussian function was assigned to the data distribution, to track the current brain state and its likelihood of entering the seizure state zone.

Results: Our results showed that we were able to identify the most and least influential nodes from LPF recordings. The influence distribution across channels was used to define a few brain states involved in epileptogenesis. With our network subspace analysis framework we can track brain-states in this feature space and show modulation towards seizure. Electrical stimulation was analyzed in this space to evaluate stimulation efficacy. Results showed that electrical stimulation is capable of influencing the current brain-state, and during correct stimulation steer the brain-state away from a seizure state and back towards a healthy or baseline state.

Conclusions: The proposed analytical algorithm provides a framework for tracking network activity in the brain that is sensitive to epileptogenic activity. This algorithm provides the first step to an analytic measure to evaluate electrical stimulation efficacy. This framework can be used by a model-based controller, in order to continuously track brain-states and deliver electrical stimulation correctly to steer the brain-state from entering an ictal state, avoiding seizures altogether.

Funding: Please list any funding that was received in support of this abstract.: This research is supported by NIH R21 NS103113. DE was supported by the Howard Hughes Medical Institute - Gilliam Fellowships for Advanced Study program.