Voice Assistant and Facial Analysis based approach to Screen Test Clinical Depression

Abstract

Depression is a medical condition that is characterized by persistent sadness, loss of interest in doing regular activities, feelings of guilt or low self-worth. It causes disturbed sleep, loss of appetite, poor concentration and an inability to work and interact. Depression is a major issue that needs to be looked into, which can otherwise lead to extreme steps such as suicide. Developing tools to aid automatic depression assessment is expected to enhance the overall health outcomes. This platform helps in detecting signs of depression, without requiring the physical presence of mental health professional. The goal of this project is to screen depression using image processing and voice assistant, which will lead to an effective system that assists psychiatrists. In this work, we automate the task of screening depression by making use of two primary aspects, namely physical signs of depression and linguistic features. The Screening process is carried out using the Patient Health Questionnaire-9 (PHQ-9) screening test. The user is categorized into one of the 5 categories based on the score: none, mild, moderate, moderately severe and Severe. Finally, based on categorization we propose a Treatment Action.