| dc.description.abstract | Electroencephalogram (EEG) data has shown great promise but requires sophisticated methods due to the complex spatial and temporal patterns found in such data, so this research was conducted with the objective to investigate the efficiency of different types of deep learning models that includes Convolutional Neural Networks (CNNs), Long Short-Term Memory(LSTMs), Hybrid CNNs and Siamese LSTMs in classifying EEG data associated with schizophrenia. What was demonstrated was that these models were able to capture the intricate patterns within EEG data exceptionally well leading to accurate predictions about the patient’s condition, and results from evaluating different models indicate that the Hybrid (CNN+LSTM) architecture offers optimal suitability for this specific application because of improved outcomes. Important implications regarding the improvement of diagnosis and treatment for schizophrenia and other neurological disorders can be achieved through deep learning models as shown by this research. | en_US |
