A convolutional neural network for the classification of Major Depressive Disorder (MDD) using Electroencephalogram (EEG) signals.
The dataset used for to train this model has been merged from two different sources:
Figure 1: EEG Waves sample of a Healthy patient |
Figure 2: EEG Waves sample of a MDD patient |
Figure 5: DeepWave Model Architecture
DeepWave achieved a training accuracy of 96.62%, training loss of 8.70%, validation accuracy of 87.05%, validation loss of 60.24%. The accuracy can be further improved by training the model for more epochs and modifying the model architecture.
Figure 6: DeepWave Accuracy Plot |
Figure 7: DeepWave Loss Plot |