Distinguishing depression from healthy controls using brain network features: A fNIRS and machine learning approach
Kechuang Zhang1, Mengbi Yang1, Min Xi2
1School of Mechanical Engineering, Northwestern Polytechnical University, Xi'an, Shaanxi, China.
Summary
Researchers identified distinct brain network connectivity patterns in depressed students using functional near-infrared spectroscopy (fNIRS). These findings suggest potential biomarkers for depression, differentiating affected individuals from healthy controls.
Area of Science:
- Neuroscience
- Psychiatry
- Biomarkers
Background:
- Depression is a prevalent mental health disorder with complex neurobiological underpinnings.
- Identifying reliable biomarkers for depression is crucial for early diagnosis and effective treatment.
Purpose of the Study:
- To explore functional near-infrared spectroscopy (fNIRS) derived local brain network connectivity as potential biomarkers for depression.
- To investigate differences in brain network features between depressed students and healthy controls.
Main Methods:
- Recruited 31 depressed students and 32 healthy controls.
- Collected resting-state and verbal fluency task (VFT) fNIRS data.
- Analyzed local connectivity features in the frontopolar region.
Main Results:
- Depressed participants showed increased frontopolar network connectivity during rest and decreased connectivity during VFT.
- Machine learning classifiers achieved AUC values over 0.7.
- The random forest model demonstrated high specificity and sensitivity during VFT.
Conclusions:
- Local brain network features identified via fNIRS show promise as biomarkers for distinguishing depression.
- These findings offer insights into the neural mechanisms of depression.
