AUTHOR=Zheng Zhiguo , Liang Lijuan , Luo Xiong , Chen Jie , Lin Meirong , Wang Guanjun , Xue Chenyang 

TITLE=Diagnosing and tracking depression based on eye movement in response to virtual reality

JOURNAL=Frontiers in Psychiatry

VOLUME=Volume 15 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2024.1280935

DOI=10.3389/fpsyt.2024.1280935

ISSN=1664-0640

ABSTRACT=Depression is a prevalent mental illness that is primarily diagnosed using psychological and behavioral assessments. However, these assessments lack objective and quantitative indices, making rapid and objective detection challenging. In this study, we propose a novel method for depression detection based on eye movement data captured in response to virtual reality (VR). We collected eye movement data to establish high-performance classification and prediction models using eXtreme Gradient Boosting (XGBoost),multilayer perceptron (MLP),SVM and Random Forest. After five-fold cross-validation, the XGBoost model had mean accuracy, precision, recall, and area under the curve (AUC) values of 76%, 94%, 73%, and 82%, respectively, and an F1score of 78%. The predicted error for the Patient Health Questionnaire-9 (PHQ-9) score ranged from -1 to 1. Similarly, the MLP model achieved classification accuracy, precision, recall, and AUC values of 86%, 96%, 91%, and 86%, respectively, and an F1-score of 92%. The predicted error for the PHQ-9 score ranged from -0.6 to 0.6. To verify the role of computerized cognitive behavioral therapy (CCBT) in treating depression, participants were divided into intervention and control groups. The intervention group received CCBT, whereas the control group did not receive any treatment. After five CCBT sessions, significant changes were observed in the eye movement indices of fixation and saccade, as well as PHQ-9 scores. Moreover, these two indices played significant roles in the predictive model. Therefore, eye-movement indices obtained using a VR eye tracker can be used as biomarkers to detect depression symptoms. Fixation and saccade indices can be used to predict depression, making them useful biomarkers. In addition, CCBT effectively treated depression.