Application of self organizing map to identify nocturnal epileptic seizures

Abstract

A patient-specific seizure detection system for Nocturnal Frontal Lobe Epilepsy (NFLE) is proposed. Data of several patients affected by NFLE, extracted from the EPILEPSIAE database, have been used for this study. As every patient possesses different physiological characteristics, several simulations were performed in order to find the best features to be extracted from electroencephalogram (EEG) signals and to be inputted to 2-dimensional Self Organizing Maps (SOM). The proposed approach allows us the definition of simple displays capable of presenting meaningful information on the actual state of the neural activities, revealing the mapping potential of clustering the data coming from seizure and non-seizure epochs; moreover, it also suggests to use SOM as seizure early detectors. In fact, the temporal sequence of the samples in an EEG recording can be projected on the SOM, obtaining a trajectory that describes the dynamics of the brain state as captured by the EEG. The analysis of the trajectory can provide information on an eventual impending seizure event. The work shows the capability of the system to early and accurately detect Nocturnal Frontal Lobe Epilepsy seizure reaching a mean value of 77.23% and 88.94% for sensitivity and specificity respectively, and highlights the possibility to promote therapies aimed at rapid and targeted treatment of seizures.