Bifurcation of Depression and Bipolar Disorder through Different Modalities using Deep Learning

People are experiencing mental health problems and abnormalities as a result of the stress and pressure that come with modern living. This paper focuses on detecting mental illnesses like Depression or Bipolar Disorder which has affected around 3.7% and 0.3% respectively, of the entire Indian Population. The majority of current methods of diagnosing psychopathology rely on verbal accounts (clinical interview or questionnaire) from patients, their loved ones, or caretakers. They lack organized and effective methods for incorporating behavioral observations that are clear signs of psychiatric disease and that frequently happen without either person's knowledge. In patients receiving treatment for depression and bipolar disorder, we contrasted the clinical diagnosis of significant depression with automatically recorded facial actions and voice prosody.