PhD Student, Maastricht University
Automated seizure detection using long-term EEG to support neurologist presurgical evaluation step
Epilepsy is a common long-term neurological condition in which nerve cell activity in the brain becomes disrupted. According to the WHO, about 0.6% of the general population has epilepsy, and nearly 80% of the affected people are found in developing regions. It manifests clearly through epileptic seizure, a period of sudden recurrent and transient disturbances of perception or behavior resulting from the excessive synchronous activity of neurons in the brain. Candidates for the surgery undergo many pre-surgery investigations where the most important is the long-term Electroencephalographic (EEG) monitoring to capture seizures for off-line analysis. A crucial step to improve the quality of life of epileptic patients is an early diagnosis so they can receive the appropriate treatments as fast as possible. Expert neurologists then visually inspect the collected EEG data to detect epilepsy-related activity, so they can determine the area where the seizures begin, called the seizure focus, and hence conclude whether the surgery is feasible.
However, locating epileptic activity in a continuous EEG recording lasting several days or weeks is an exhausting, demanding and time-consuming task because such activity represents a small percentage of the entire recording. These difficulties have motivated the development of automated methods that scan, identify, and then present to a neurophysiologist epoch containing epileptic events. Such systems help to overcome the limitations of the traditional visual inspection performed by expert neurologists and avoid any misreading or missing information.
Our proposed study suggests a novel parameter, referred as power band index (PBI) derived from EEG oscillations rhythms for real-time seizure detection. The classification (normal or epileptic) was performed using the proposed PBI with a threshold function derived from initial 10s duration of normal PBI segments. The algorithm was tested on 89 hrs of EEG with 66 seizures from 10 subjects. Results revealed that mean sensitivity of 95.8% and mean false detection rate of 0.38 /hr, and mean detection delay of 1.1s. The proposed scheme was found to be patient independent, training free and less computationally intensive compared to other known real-time algorithms reported earlier.