Detecting Schizophrenia through EEG signals using Deep Learning

Millions of people between the ages of 16 to 30 are diagnosed with Schizophrenia every year. Schizophrenia is a serious mental disorder characterized by abnormalities in the individual’s cognition, emotional as well as physical responses. This study examines the detection of schizophrenia using electroencephalogram (EEG) signals collected from the human brain. EEG signals provide valuable information about cognitive responses of an individual and have the potential to serve as a biomarker for diagnosing schizophrenia. This study analyzes a large dataset of EEG signals from people with and without schizophrenia using advanced machine learning and signal processing techniques. We aim to demonstrate a robust and accurate classification model for distinguishing schizophrenic patients from those not affected by it, based on data recorded through Electroencephalogram signals. Various features are extracted from EEG signals, including frequency domain measurements, event-related potentials, and connectivity patterns. These features are then used to train and evaluate various machine learning as well as deep learning algorithms such as convolutional neural networks (CNNs), Ensemble Methods, Support Vector Machines (SVMs), etc. Performance of the deep learning model was evaluated using various statistical metrics like accuracy, sensitivity, specificity, and the area under receiver operating curve (AUC-ROC). It is found that the potential of EEG-based methods for early schizophrenia detection and monitoring, improving clinical diagnosis and treatment strategies.