A One-Dimensional Convolutional Neural Network Model for Automated Localization of Epileptic Foci

Abstract

Intracranial electrocorticogram (iEEG) is often used by clinical experts to determine the location of the epileptic focal in the treatment of epilepsy. However, assess the location of epileptic foci by using iEEG is time-consuming and strenuous for clinical experts. Technology for automated localization of the channel of epileptic focal is indispensable. Hence, we developed a one-dimensional convolutional neural network (ID-CNN) model, which can directly extract features and train model by the raw signals without preprocessing, and performed the classification of focal and nonfocal epileptic iEEG signals. Compared with other machine learning methods, the amount of parameter reduced significantly. Our developed model has yielded the classification accuracy of 85.14% in classifying the focal and nonfocal epileptic iEEG signals.