CLASSIFICATION OF EPILEPTIC SEIZURE USING A ONE-DIMENSIONAL CONVOLUTIONAL NEURAL NETWORK

Technological advances in artificial intelligence have enabled scientists to obtain significant data in different application areas. Pattern recognition tasks can now be effectively performed on large datasets using artificial intelligence algorithms, particularly machine learning and deep learning techniques. In the context of epilepsy a neurological disorder electroencephalogram (EEG) signals are widely utilized to gather information about the brain’s electrical activity. In this study, the performance of five machine learning algorithms and one deep learning model was evaluated for the classification of epileptic seizures versus normal conditions using EEG signals obtained from individuals diagnosed with epilepsy. The machine learning techniques employed included K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Adaptive Boosting (AdaBoost), Gaussian Naive Bayes (GNB), and Random Forest (RF), while the deep learning approach was based on a one-dimensional Convolutional Neural Network (1D-CNN). The dataset used for model training and evaluation was the publicly available University of Bonn EEG dataset. Among the machine learning methods, the Random Forest classifier achieved the highest performance, with an accuracy of 0.96, recall of 0.89, precision of 0.91, and an F1-score of 0.90. The 1D-CNN model demonstrated comparable results, achieving an accuracy of 0.96, recall of 0.87, precision of 0.93, and an F1-score of 0.90. These findings indicate that while the deep learning model provided a marginal improvement in precision over the best-performing machine learning algorithm, both approaches yielded similarly high classification performance in the detection of epileptic seizures from EEG data.