Enhanced major depressive disorder diagnosis: A hybrid deep learning framework utilizing EEG and fMRI

Major Depressive Disorder (MDD) is a severe mental disorder that requires early detection, which poses a need to enhance diagnostic accuracy. Although Electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI) have demonstrated the ability to diagnose MDD, their combined application has not been extensively investigated. This work presents a new two-stream deep learning model for MDD diagnosis with EEG and fMRI data acquired simultaneously. The proposed approach is to process EEG and fMRI data independently by two different Convolutional Neural Networks (CNNs) to obtain discriminative features simultaneously. These features were merged and then classified using a support vector machine optimized by a genetic algorithm. The proposed model achieved a mean diagnostic accuracy of 98.67%, along with a recall of 96.43% and a precision of 100%. This study demonstrates the framework’s effectiveness as a tool to assist clinicians in enhancing the accuracy of MDD diagnosis and facilitating timely treatment planning.