Stochastic Systems Group
Home Research Group Members Programs  
Demos Calendar Publications Mission Statement Alumni

SSG Seminar Abstract

Graphical Models for Guiding Neurosurgery in Epileptic Patients

Justin Dauwels

For approximately 30% of epilepsy patients, seizures are poorly controlled with medications alone. For those patients, surgery may be an option: the aim is to remove the brain area(s) where the seizures originate. Such surgeries are nowadays carried out routinely, often with considerable success. The key to success is to be able to accurately localize the seizure onset zone. In order to delineate that zone, one heavily relies on brain signals recorded during seizures (ictal recordings), by electrodes that are semi-chronically implanted in the patient’s brain. Since seizures usually occur only occasionally, patients often have to stay several weeks to even months in the hospital until sufficient seizures have been recorded. Obviously, this procedure is costly, uncomfortable, and not without risk for infection and other side effects.

The objective of our research is to substantially shorten the hospitalization of epilepsy patients, from several weeks to a few days at most. The underlying idea is to exploit brain signals recorded while the patients do not suffer from seizures (interictal recordings). We have developed statistical inference methods that extract and combine various characteristics of interictal recordings in order to localize the seizure focus. At the core lies an Ising model that imposes spatial continuity. For the 11 patients considered so far, our algorithms are able to reliably localize the seizure focus (as determined by clinicians from ictal EEG).

In the long term, this research project may enable shorter hospitalization or even avoidance of semi-chronic implantations altogether.

This is joint work with Dr. Sydney Cash, M.D.(MGH/Harvard).

Problems with this site should be emailed to