Related Experiment Video
Updated: Aug 16, 2025
05:19
Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
Published on: July 7, 2023
2.4K
Few-Electrode EEG from the Wearable Devices Using Domain Adaptation for Depression Detection
Wei Wu1,2, Longhua Ma1, Bin Lian1
1School of Information Science and Engineering, NingboTech University, Ningbo 315100, China.
Biosensors
|December 23, 2022
Summary
Domain adaptation using deep learning improves electroencephalogram (EEG) analysis for detecting major depressive disorder (MDD). This method enhances accuracy by minimizing data variations between individuals, leading to more reliable depression diagnosis.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Medical Diagnostics
Background:
- Major Depressive Disorder (MDD) is a significant global health concern.
- Electroencephalogram (EEG) signals show promise for depression detection.
- Variability and non-uniformity in EEG data across subjects hinder accurate diagnosis.
Purpose of the Study:
- To apply deep learning with domain adaptation for improved MDD detection using EEG signals.
- To address the challenges posed by time-varying and non-uniform EEG data distributions.
- To enhance the accuracy and reliability of depression diagnosis from EEG.
Main Methods:
- EEG signals were preprocessed.
- EEG data were transformed into image representations using two methods: separate channel display and RGB synthesis.
- A domain adaptation deep learning model was employed for training and prediction.
Main Results:
- The domain adaptation model effectively extracted relevant EEG features.
- An average accuracy of 77.0 ± 9.7% was achieved in depression detection.
- The method demonstrated a significant reduction in inherent EEG signal differences.
Conclusions:
- Domain adaptation is effective in mitigating inter-subject variability in EEG data.
- This approach leads to more accurate and reliable diagnosis of depression across different users.
- Deep learning with domain adaptation offers a promising avenue for objective MDD assessment.
