Survey on Artificial Intelligence- based Depression Detection using Clinical Interview Data

Depression is a current and significant mental illness that most people face in their daily lives. Depression affects not just the emotional state of the person but also the physical, psychological and mental state of the individual. In contrast to other diseases, depression cannot be detected by laboratory tests, remains undetected due to a lack of understanding and ignorance, and can worsen, leading to suicide. Currently, the diagnostic procedure for detecting depression adopts face-to-face interactions and self-reported questionnaires by physicians. Recognizing depression necessitates a psychiatric analysis of human behavior and social interactions. People with depression behave differently than normal people, as can be seen in the patients’ audio and video recordings. Computer vision and machine learning have been used to detect depression to fight the disease at an earlier stage. This article introduces a survey of research on the identification of depression using clinical interview data through Unimodal and Multimodal approaches.