Current Research:
Epilepsy is one of the most common neurological disorders. It affects about 50 million people worldwide, 2.7 million in the U.S. alone. Over 180,000 Americans develop epilepsy each year. Epilepsy is seriously debilitating in about 30% of these patients because their seizures cannot be adequately controlled with current medications (e.g. Kwan and Brodie, 2000). Many intractable seizure patients are children who will require treatment and care for their whole lives, causing a significant burden on their families and our health care system. Despite the human and financial costs of epilepsy, its underlying causes remain poorly understood, and effective treatments are not available for a significant number of patients.
Our work concentrates on three major questions in epilepsy. (1) Where is seizure activity located in the brain? (2) When do seizures occur? (3) How do seizures start, propagate, and stop? Each of these questions has specific scientific and clinical relevance. The answer to the localization question (where?) may ultimately guide surgical resection procedures in epilepsy. An algorithm to detect seizures (when?) is currently used by clinicians to find seizure activity in multi-day long-term monitoring records. In addition, detection or knowledge of immanent seizure occurrence may guide our research by highlighting specific epochs of an EEG record to look for onset processes, and ultimately focus therapeutic intervention before or at the onset of seizures. The third question concerning the mechanism underlying seizure activity (how?) is of principal interest for neurologists and clinical neurophysiologists, as medication effects are commonly described at the molecular/cellular level, while clinical effects are evaluated at organ/organism scales.
We have developed both computational and experimental models to study seizure-like activity. Brain slices from patients undergoing resection surgery to treat their epilepsy are studied with electrophysiological and (recently) optical techniques. These same techniques are also applied to study an in vitro seizure model of mouse neocortex. Our published findings show that both the level of synchrony and synaptic excitation during seizure activity may very well be significantly lower than previously thought and the role of persistent sodium channels in the generation of network bursting (e.g. van Drongelen, Koch, Marcuccilli, Peña, and Ramirez, 2003; van Drongelen, Lee, Hereld, Jones, Cohoon, Elsen, Papka, and Stevens, 2004; van Drongelen, Lee, Koch, Hereld, Elsen, Chen, and Stevens, 2005; van Drongelen, Koch, Elsen, Doren, Lee, Marcuccilli, Hereld, Stevens, and Ramirez, 2006; van Drongelen, Lee, Stevens and Hereld, 2007). These findings are important for understanding bursting activity in neuronal populations and for the development of treatment for seizures. Recently this work was featured as one of the major new findings in a review paper on modeling in epilepsy published in Nature Reviews Neuroscience (9:626-637; 2008).
The theoretical approach includes computational modeling and mathematical modeling. Computer modeling is done in collaboration with Dr. Hereld from the Computation Institute and Argonne National Laboratory, Argonne, IL and Dr. Pesce from the Computation Institute. Computer resources to support our extensive modeling effort are made available by the Division of Mathematics and Computer Science, Argonne National Laboratory, Argonne, IL and The Computation Institute.
Personnel
Ivan Goussakov, PhD
Mihailo Radojicic, MD
Michael Carroll, PhD
Albert Wildeman |
