A Pilot study for Depression Detection during COVID-19 using Stroop Test

The Corona virus disease 2019 (COVID-19) pandemic has caused substantial increase in distress among people all over the world. This work aims to study depression during the COVID-19 among the educational sector and to develop a novel stroop test based depression detection system by analyzing the response time (RT) for normal stroop test (ST) and emotional stroop test (EST). The data for this work is collected from 44 participants. It is found that 66% of the participants have depression. The analysis of RT for ST and EST before and after showing video stimulus, indicates that there is a significant difference in change of response time (dRT) for both normal and depressive cases. Further this feature along with the physiological parameters (PhyP) of the participants are given to support vector machine (SVM) and extreme gradient boost (XGBoost) classifier to develop depression detection systems. The XGBoost provides highest accuracy of 85.71% with PhyP + ST_dRT data and an accuracy of 71.43% with PhyP + EST_dRT data. Therefore, the proposed systems may serve as a screening tool for depression during this pandemic situation.