The outcome for patients undergoing epilepsy surgery may be improved with the use of computational methods to identify more precisely the origin of seizures in the brain.
Research published in Neurology explains modern computational methods mean there are more detailed ways in which we analyse EEG (a device which gathers information about brain activity and seizures using electrodes placed on the scalp) signals to identify the area of the brain where seizures originate.
In a collaboration with other researchers from the UK and USA, researchers from the UCL Institute of Neurology, led by BRC supported Professor Matthew Walker, analysed data collected from patients who already had EEG signals recorded.
Professor Walker said: “The computational analysis method is to try and extract greater information from EEG recordings. Rather than just looking at where we think the seizure may be beginning, we analysed the very high frequency EEG signal as the seizures begin, to see where that area is and how the seizure is spreading.”
The patients taking part in this study have drug resistant epilepsy. With patients of this kind the best option left to cure their epilepsy is to surgically remove the part of the brain that is causing the seizures. The challenge is to identify which part of the brain the seizures are coming from. If there is insufficient precision, the surgery may result in neurological deficits and if the wrong areas are selected, the epilepsy does not improve.
The deep analysis of the EEG signals provides a much better idea of the parts of the brain that would need to be removed in order for a patient to be free of seizures. It can also point out the exact areas of the brain that should be removed, potentially resulting in less extensive surgery for many patients.
Professor Walker said: “The clinical implications are twofold: firstly, we are going to be able to do less extensive surgery, causing less of a deficit. Secondly, we are going to have a greater success with that surgery. Surgery for epilepsy cures, at best, 70% of patients. This is likely to be less than 50% for the types of epilepsy in this study. These methods have the potential to improve those figures dramatically.”
To read Seizure localization using ictal phase-locked high gamma click here.
