Resting-state EEG Classification of Children and Adolescents Diagnosed with Major Depression Disorder Using Convolutional Neural Network

Jahanian Najafabadi A, Bagh K. Resting-state EEG Classification of Children and Adolescents Diagnosed with Major Depression Disorder Using Convolutional Neural Network. 2023.Previous studies have applied Deep learning models on Electroencephalography data for classification and prediction tasks. In this work, we aim to test a model using convolutional neural network (CNN) in order to identify biomarkers in childhood and Adolescents with major depression disorder from age matched healthy young individuals. CNN used to transfer learning from the VGG16 network. The data were tested using several filtered frequencies combined with a preprocessing pipeline. Overall, results achieved an accuracy of 0.875 and F1-Score of 0.638 and revealed that while lower Delta (0.856 accuracy and 0.539 F1-Score), and higher Theta activity (0.895 accuracy and 0.717 F1-Score), and higher Alpha (0.804 accuracy and 0.606 F1-Score) were classified in MDD compared with healthy. We did not find whether Beta and Gamma are biomarkers for MMD with higher accuracy. We further showed that while in MDD group, delta frequency bands were featured in left temporal, occipital, bilateral frontal and central regions; theta frequency bands were featured in left temporal and frontal, left occipital and central regions, and Alpha frequency band was featured in left frontal, central, left occipital and left temporal regions