Computer assisted identification of stress, anxiety, depression (SAD) in students: A state-of-the-art review

Stress, depression, and anxiety are a person's physiological states that emerge from various body features such as speech, body language, eye contact, facial expression, etc. Physiological emotion is a part of human life and is associated with psychological activities. Sad emotion is relatable to negative thoughts and recognized in three stages containing stress, anxiety, and depression. These stages of Physiological emotion show various common and distinguished symptoms. The present study explores stress, depression, and anxiety symptoms in student life. The study reviews the psychological features generated through various body parts to identify psychological activities. Environmental factors, including a daily routine, greatly trigger psychological activities. The psychological disorder may affect mental and physical health adversely. The correct recognition of such disorder is expensive and time-consuming as it requires accurate datasets of symptoms. In the present study, an attempt has been made to investigate the effectiveness of computerized automated techniques that include machine learning algorithms for identifying stress, anxiety, and depression mental disorder. The proposed paper reviews the machine learning-based algorithms applied over datasets containing questionnaires, audio, video, etc., to recognize sad details. During the review process, the proposed study found that artificial intelligence and machine learning techniques are well recommended and widely utilized in most of the existing literature for measuring psychological disorders. The various machine learning-based algorithms are applied over datasets containing questionnaires, audio, video, etc., to recognize sad details. There has been continuous monitoring for the body symptoms established in the various existing literature to identify psychological states. The present review reveals the study of excellence and competence of machine learning techniques in detecting psychological disorders' stress, depression, and anxiety parameters. This paper shows a systematic review of some existing computer vision-based models with their merits and demerits.