New Applied Mathematics and Nonlinear Sciences Study Results from North China University of Water Resources and Electric Power Described [Automatic identification of depressive symptoms in college students: an application of deep learning-based...].

A recent study conducted by researchers at North China University of Water Resources and Electric Power explores the use of deep learning-based Convolutional Neural Networks (CNN) to automatically identify depressive symptoms in college students. The study focuses on analyzing facial behavior as a direct and easily accessible source of behavioral data. The researchers found that depressed patients exhibit unique behavioral patterns, including reduced positive emotional feedback, enhanced negative emotional feedback, misjudgment of neutral stimuli as negative stimuli, and slow changes in expression behavior. The CNN-LSTM model achieved a recognition accuracy of 73.21% and a recall rate of 85.71% when applied, making it suitable for primary screening of depression in college students. This research provides a methodological basis and technical support for the automatic identification of depressive symptoms. [Extracted from the article]