Related Experiment Video
Updated: Jul 2, 2025
07:04
Virtual Reality Tools for Assessing Unilateral Spatial Neglect: A Novel Opportunity for Data Collection
Published on: March 10, 2021
4.0K
Diagnosing and tracking depression based on eye movement in response to virtual reality
Zhiguo Zheng1,2, Lijuan Liang3, Xiong Luo4
1School of Information and Communication Engineering, Hainan University, Haikou, China.
Frontiers in Psychiatry
|February 20, 2024
Summary
This study introduces a novel method for detecting depression using virtual reality (VR) eye-tracking data, identifying fixation and saccade as key biomarkers. Computerized cognitive behavioral therapy (CCBT) also showed effectiveness in improving depression symptoms.
Area of Science:
- Neuroscience
- Psychiatry
- Computer Science
Background:
- Depression diagnosis relies on subjective assessments, lacking objective and quantitative measures.
- Current diagnostic methods present challenges for rapid and objective detection of depression.
- Virtual reality (VR) offers a novel platform for objective data collection in mental health assessments.
Purpose of the Study:
- To propose and validate a new method for depression detection using eye movement data captured via VR.
- To evaluate the efficacy of machine learning models in classifying depression based on eye movement patterns.
- To investigate the potential of eye movement indices as biomarkers for depression and assess the impact of computerized cognitive behavioral therapy (CCBT).
Main Methods:
- Collected eye movement data from participants using a VR eye tracker.
- Developed and evaluated four machine learning models (XGBoost, MLP, SVM, Random Forest) for depression classification.
- Assessed model performance using five-fold cross-validation and metrics like accuracy, precision, recall, AUC, and F1-score.
- Analyzed changes in eye movement indices (fixation, saccade) and Patient Health Questionnaire-9 (PHQ-9) scores before and after CCBT intervention.
Main Results:
- The Multilayer Perceptron (MLP) model achieved the highest classification accuracy (86%) with an F1-score of 92%.
- Eye movement indices, specifically fixation and saccade, demonstrated significant roles in predicting depression symptoms.
- Participants undergoing CCBT showed significant improvements in eye movement indices and PHQ-9 scores compared to the control group.
Conclusions:
- Eye movement indices captured via VR eye tracking can serve as effective biomarkers for depression detection.
- The findings support the use of VR-based eye tracking as an objective tool for mental health assessment.
- Computerized cognitive behavioral therapy (CCBT) is an effective treatment for depression, as indicated by changes in eye movement patterns and symptom scores.
