Early Study in Automatic Identification of Epilepsy in Neonatal Using EEGLAB and One Dimensional Convolutional Neural Network Through the EEG Signal

In detecting epileptic activity, medical experts examine the visual result of Electroencephalography signals. The analysis will take a lot time and effort, due to large amount data. Furthermore, there are some errors in concluding result. One ways analyze this quickly is use Machine Learning (ML) methods. This study aims evaluate performance 1D-CNN identifying given First, signal go through pre-processing using EEGLAB Toolbox which then classified identify epilepsy non-epilepsy with algorithm. results showed that proposed method obtained high accuracy values, respectively 99,078% for training data 82,069% validation results. From evaluation by confusion matrix, an average 99,31% was obtained. Based on evaluation, model can be used as efficient process automatic classification, detection, or identification activity.