Related Experiment Video
Updated: May 13, 2026
08:51
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
Published on: November 1, 2019
5.6K
Evaluating Augmentation Approaches for Deep Learning-based Major Depressive Disorder Diagnosis with Raw
Charles A Ellis1, Robyn L Miller1, Vince D Calhoun1
1Center for Translational Research in Neuroimaging and Data Science Georgia State University, Emory University, Georgia Institute of Technology Atlanta, USA.
Biorxiv : the Preprint Server for Biology
|January 8, 2024
Summary
Data augmentation for electroencephalography (EEG) in major depressive disorder diagnosis shows limited benefits. Channel dropout augmentation improved model performance, unlike other methods when compared to duplicated data baselines.
Area of Science:
- Neuroscience
- Machine Learning
- Biomedical Engineering
Background:
- Deep learning models show promise for neuropsychiatric disorder diagnosis but are limited by small datasets.
- Data augmentation (DA) is a potential solution, yet its effectiveness for electroencephalography (EEG) in neuropsychiatric disorders is underexplored.
Conclusions:
- Comparing EEG DA methods against a duplicated data baseline of equal size is crucial.
- Channel dropout is a promising DA technique for EEG-based MDD diagnosis.
- Findings provide insights for developing effective deep learning strategies for small EEG datasets in neuropsychiatry.
