image image


A comprehensive mental onset detection system tailored for the needs of universities and colleges with a focus on the mental health of students.

  • Forecasting depression through a multimodal strategy.

  • Offering an AI companion for individuals to discuss their struggles with.

  • Providing a comprehensive evaluation and visual analysis of students on admin dashboard


  • Regular updated mental health data of students
  • Insightful mental health data analysis
  • Custom cross platform mobile application for students
  • Powered by numerous machine learning models (ensemble learning)

System Diagrams

  • Overall image


The app is built with Flutter.

The app takes in the user video interview, quizzes and various other data to predict the mental health of the user.


This is built with NextJS.

This is the web dashboard for the university/college to view the mental health data of the students.

The dashboard has a variety of features such as:

  • View aggregate mental health data of students
  • View mental health data of students in a particular course
  • Classify trends based on branch, age, gender

Machine Learning and Backend

The machine learning and backend is built with Python and Flask.

The machine learning diagram:


Product Demos


  1. DASS 21 – Questionnaire: Queensland Mental Health Commission. (n.d.). Depression, Anxiety and Stress Scale (DASS 21). https://maic.qld.gov.au/wp-content/uploads/2016/07/DASS-21.pdf

  2. DAIC-WOZ Database: This database contains clinical interviews designed to support the diagnosis of psychological distress conditions such as anxiety, depression, and post-traumatic stress disorder https://dcapswoz.ict.usc.edu/


View Github