Related Experiment Video
Updated: Jan 13, 2026
Polar Histogram Visualization of Acute Stress Disorder Scale Scores for Comprehensive Clinical Assessment
Published on: December 6, 2024
Machine Learning Based Identification of Depressive Symptoms Among Students in a Chinese University Using Functional
Yange Wei1, Yuanle Chen1, Ning Wang2
1Department of Early Intervention, Mental Health and Artificial Intelligence Research Center, The Second Affiliated Hospital of Xinxiang Medical University, Henan Mental Hospital, 453002 Xinxiang, Henan, China.
Functional near-infrared spectroscopy (fNIRS) combined with a verbal fluency task (VFT) shows promise for objectively detecting depression in university students. This method identified specific brain activity patterns linked to depression, particularly in the left medial prefrontal cortex.
Area of Science:
- Neuroscience
- Clinical Psychology
- Biomedical Engineering
Background:
- University students experience high rates of depression, necessitating objective diagnostic tools beyond subjective questionnaires.
- Current depression assessment methods are often subjective, highlighting the need for reliable, objective measures.
- Functional near-infrared spectroscopy (fNIRS) offers a potential avenue for objective depression detection.
Purpose of the Study:
- To investigate functional near-infrared spectroscopy (fNIRS) signal changes associated with depression in university students.
- To assess the efficacy of fNIRS signals in the automatic detection of depression.
- To explore the neural correlates of depression using fNIRS during a cognitive task.
Main Methods:
- 192 university students underwent psychological assessment and fNIRS measurement during a verbal fluency task (VFT).
- Cerebral blood oxygenation signals were recorded using a 48-channel fNIRS system.
- Machine learning classifiers, including K-Nearest Neighbors (KNN), were employed to differentiate depression using fNIRS data, with performance evaluated via ROC curves and AUC.
Main Results:
- Significant hemodynamic differences were observed in the depression group compared to controls, particularly in the medial prefrontal cortex (MPFC) and left temporal lobe.
- The left MPFC, left dorsolateral prefrontal cortex, and left temporal lobe showed associations with depression.
- The KNN classifier achieved the highest performance (AUC = 66.51%), with the left MPFC being a key predictor.
Conclusions:
- fNIRS combined with VFT presents a viable objective method for assessing depressive symptoms in university students.
- The findings emphasize the critical role of the left MPFC in the neurobiology of depression.
- An fNIRS-based system for identifying depression in university students was successfully developed.
