Journal
JOURNAL OF PSYCHIATRIC RESEARCH
Volume 161, Issue -, Pages 377-385Publisher
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.jpsychires.2023.03.023
Keywords
Machine learning; Spectral analysis; Cortical oscillation; Precision psychiatry; Electroencephalogram
Categories
Ask authors/readers for more resources
Major depressive disorder has a significant impact on global healthcare resources. Antidepressants are commonly used as the first-line treatment, but brain stimulation therapy may be necessary for non-responders. Digital phenotyping can help predict treatment effectiveness for patients with major depressive disorder.
Major depressive disorder places a great burden on healthcare resources worldwide. Antidepressants are the first -line treatment for major depressive disorder, but if patients don't respond adequately, brain stimulation therapy may be needed as second-line treatment. Digital phenotyping in patients with major depressive disorder will aid in the timely prediction of treatment effectiveness. This study explored electroencephalographic (EEG) signatures that diversify depression treatment responsivity, including antidepressant administration or brain stimulation therapy. Resting-state, pre-treatment EEG sequences from depressive patients who received fluoxetine treatment (n = 55; 26 remitters and 29 poor responders) or electroconvulsive therapy (ECT, n = 58; 36 remitters and 22 nonremitters) were recorded on 19 channels. Twenty-nine EEG segments were obtained from each patient per recording electrode. Power spectral analysis was conducted for feature extraction and showed the highest pre-dictive accuracy for fluoxetine or ECT outcomes. Both occurred with beta-band oscillations within right-side frontal-central (F1-score = 0.9437) or prefrontal areas of the brain (F1-score = 0.9416), respectively. Signifi-cantly higher beta-band power was observed among patients who lacked adequate treatment response than the remitters, specifically at 19.2 Hz or 24.5 Hz for fluoxetine administration or ECT outcome, respectively. Our findings indicated that pre-treatment, right-side cortical hyperactivation is associated with poor outcomes of antidepressant-based or ECT-based treatment in major depression. Whether depression treatment response rates can be improved by reducing the high-frequency EEG power in corresponding areas of the brain to provide a protective effect against depression recurrence warrants further study.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available