Computational Neurophysiology

Our research on computational neurophysiology is focused on improving our understanding of epilepsy and the control of seizures. In our work on epilepsy, we seek to employ our knowledge of seizure generation, sophisticated continuous brain sensing methods, and advanced computational methods for seizure forecasting and localizing the seizure onset area.

The unpredictability of seizures is one of the most disabling aspects of epilepsy. In this project we seek to forecast periods when seizures are more likely, by tracking changes in brain electrical activity, mood, stress, sleep, and body chemicals in individuals who have electrodes implanted in their brain for treatment of their epilepsy. There are daily and multi-day cyclical vulnerabilities to seizures and observable neurochemical and behavioral correlates which underlie these cyclical vulnerabilities. Our goal is to eventually control seizures in individuals with medically refractory epilepsy and to develop seizure forecasting in individuals with the RNS System to then extend to all individuals with epilepsy.

Accurate localization of the seizure onset area in patients with medically refractory focal epilepsies is crucial for successful neurosurgical treatment using interventions such as tissue resection, laser ablation and brain stimulation. We have developed brain interictal network mapping algorithm, by which a brief segment of the brain’s interictal (background) activity is used to first establish a frequency band-specific functional connectivity map and then to accurately delineate the seizure onset area in focal epilepsies. Our objective is to delineate the seizure onset area more effectively and accurately than current approaches and reduce the cost and patient discomfort and improve the outcome of epilepsy surgery.