Major depressive disorder diagnosis from electroencephalogram data and potential treatment with dimethyltryptamine
International Advanced Researches and Engineering Journal August 15, 2023 Peer reviewed DOI: 10.35860/iarej.1231288
Summary
The study analyzed resting-state EEG signals from patients with Major Depressive Disorder (MDD), dimethyltryptamine (DMT) users, and healthy controls to evaluate a computer-aided approach for diagnosis. It found that the EEG signal power in various brain areas was significant, particularly among DMT users compared to MDD individuals and healthy controls. This suggests that DMT may stimulate brain activity in regions typically underactive in MDD patients.
Study at a glance
| Population | patients with Major Depressive Disorder, DMT users, and healthy controls |
|---|---|
| Key finding | The EEG signal power from frontal, temporal, parietal, and occipital brain areas was found to be significant in distinguishing between MDD patients and DMT users. |
Abstract
In recent times, there has been increasing interest in utilizing EEG-based techniques for studying Major Depressive Disorder as a dynamic method. Although it is frequently used for identifying depression, the method is still difficult to interpret. The conventional treatment of MDD involves medications such as Selective Serotonin Reuptake Inhibitors, which often have adverse effects. On the other hand, the use of dimethyltryptamine to stimulate brain activity in regions where MDD patients show lower activity has demonstrated promising results. This study analyzed resting-state EEG signals from MDD patients, DMT users, and healthy controls to evaluate and validated a computer-aided approach. The brain activity of DMT users was recorded and compared with MDD individuals and healthy controls. Using Welch's method, the power of several frequency bands was analyzed from the EEG dataset for comparison and diagnosis. The extracted EEG data underwent noise removal and feature extraction. The features from all controls were concatenated to form a data matrix. Furthermore, the data matrix was standardized using the Z-score standardization method. The classifier model logistic regression was employed to train and test the extracted features. The results of the investigations have demonstrated the most important features, such as signal power of the EEG data from the frontal, temporal, parietal, and occipital brain areas, to be significant.