Analysis of Epileptic Seizures Using 1D Convolutional Neural Network for Epilepsy Prediction
摘要
Epilepsy is a neuronal disease where electrical discharges occur at focal or general level on the surface of the brain resulting in involuntary seizures of the patient and especially physical damage such as bites during the attack. These electrical signals can be obtained through encephalograms, a neurophysiological test, which record the electrical activity of the brain by means of electrodes. The methodology used consisted of taking these EEG signals and passing them through bandpass filters as signal pre-processing. Subsequently, their characteristics were analyzed and extracted to detect a possible early epileptic seizure through a neural network to classify the EEG signals, and finally their training and evaluation of the classifiers. For the development of the research work, Matlab software was used for signal filtering, in addition to Google Colab where the neural network was developed together with the training of the network and a database of electroencephalographic signals.