BOB: Mental Health and Psychological Support System

A web-based mental health support platform for college and university students. The project is designed to provide a private, accessible space where students can track their mood, complete a DASS-21 assessment, use self-care tools, speak with the local BOB chatbot, and book counselling appointments.

The work is based on the dissertation topic "Development of a Mental Health and Psychological Support System", which focuses on stigma-free student support, confidentiality, local deployment, and timely access to counsellors.

Why This Project

This project was built to reduce the gap between students and mental health support on campus. Many students avoid seeking help because of stigma, privacy concerns, or long counselling wait times. BOB provides a private digital space where students can take early steps toward support through self-care tools, mood tracking, assessments, chatbot guidance, and appointment booking.

Features

• Student registration and login with JWT-based cookie sessions
• Password hashing with Bcrypt
• Mood logging with daily updates, history, streaks, and weekly statistics
• DASS-21 assessment with depression, anxiety, and stress severity scoring
• Counselling appointment booking for authenticated students
• Counsellor dashboard for viewing, completing, and cancelling appointments
• User profile page with appointment, mood, and DASS-21 history
• Wellness section with breathing exercises, wind-down activities, and curated videos
• BOB chatbot integration through a local Ollama endpoint
• Discord/community entry point from the home page

Tech Stack

• Frontend: React 19, Vite, Tailwind CSS 4, Material UI
• Backend: Node.js, Express, Prisma ORM, PostgreSQL
• Language: TypeScript across frontend and backend
• Auth & Security: JWT, bcryptjs, cookie-parser, CORS
• AI Integration: Ollama local API with bobbuddy:latest
• Utilities: axios, react-toastify, react-hot-toast
• UI & Components: lucide-react, react-icons, Tailwind CSS

Project Structure

bob/
  front/
    src/
      components/        Reusable UI, chatbot, wellness, home, and layout components
      pages/             App pages such as Home, Login, DASS21, Appointment, Dashboard

  back/
    src/
      app.ts             Express app setup, middleware, and route mounting
      server.ts          API server entry point
      routes/            Auth, appointment, counsellor, mood, DASS, profile, chat routes
      controllers/       Request handlers and business logic
      middlewares/       Authentication middleware
      config/            Prisma client configuration
    prisma/
      schema.prisma      PostgreSQL data model
      migrations/        Database migration history

Prerequisites

• Node.js
• PostgreSQL
• npm
• Ollama, if using the chatbot feature

Setup and Run

Install backend dependencies and prepare Prisma:

cd back
npm install
npx prisma migrate dev
npm run dev

Install frontend dependencies and start Vite:

cd front
npm install
npm run dev

The frontend runs on http://localhost:5173 by default, and the backend runs on http://localhost:5000 unless PORT is changed.

Chatbot Setup

The backend sends chat requests to Ollama using the model name bobbuddy:latest. Make sure Ollama is running locally and that the model is available before using the chat page.

ollama serve

Main API Areas

• POST /auth/signup and POST /auth/login
• GET /auth/session and POST /auth/logout
• POST /mood/log, GET /mood/history, GET /mood/stats
• POST /dass/result, GET /dass/my-results
• POST /book/appointment, GET /book/my-appointments
• GET /counsellor/dashboard, GET /counsellor/stats
• POST /chat/message

Build

cd back
npm run build

cd ../front
npm run build

Notes

This platform is intended as a support and early-intervention tool for student wellbeing. It is not a replacement for professional medical care or formal clinical diagnosis. The dissertation describes a privacy-first system with local deployment as a key goal; this repository implements the web application, API, database schema, and local Ollama chatbot integration needed for that system.