Epileptic Seizure Analysis using Scalp EEG Signals with Deep Learning

There has been an increase in interest in using Electroencephalogram (EEG) data for emotion detection in recent years, with possible applications in industries including affective computing and medicine. It has been discovered that persons with depression, a common mental health disorder, have differential spatial responses to neurophysiological signals when exposed to positive and negative stimuli that differ from those in healthy people. Our investigation contrasts previous research in the field and focuses on important elements essential to emotion recognition, such as themes, extracted features, classifiers, and more. This research study develops an emotion activation curve to more clearly depict the activation process of emotions. The proposed study employs machine learning to classify emotions by extracting features from EEG signals. It leverages data from several experiments to train a suggested model, which is then tested for its effectiveness in detecting emotions. The main goal of our project is to identify emotions from brain signals by applying a novel adaptive channel selection method that can distinguish between how brain activity operates in various people and emotional states.The results of our investigation show a significant improvement in the proposed method’s ability to correctly categorize positive or negative emotions in depressed patients.