MediAsk - Health Q&A for Workers

Live Demo: hackathon-ioqp.onrender.com

MediAsk is a community health Q&A platform designed for factory workers, food processors, and manual labourers in the UK. It combines NHS-verified guidance, real worker experiences, and AI-powered responses to answer health questions that matter most to working people.

Why MediAsk?

Factory workers face unique health risks - back injuries from heavy lifting, cold exposure in processing plants, chemical burns, hearing damage, and mental health struggles from isolation and shift work. Most health platforms don't address these workplace-specific issues. MediAsk does.

• 70% of factory workers report musculoskeletal pain
• 40% of shift workers suffer from sleep disorders
• Many immigrant workers don't know their NHS rights or workplace protections

Features

For Workers

• Ask Health Questions - get answers from NHS sources, experienced workers, and AI
• Voice Input - speak your question instead of typing (useful with gloves/dirty hands)
• Multi-Language Support - Google Translate with 18 languages (Hindi, Polish, Romanian, Urdu, Bengali, Arabic, and more)
• Workers Health Hub - dedicated section for factory-specific issues (back pain, cold rooms, RSI, chemicals, shift work, legal rights)
• NHS Health Lookup - search NHS conditions database in real-time
• Hazard Reporting - report unsafe conditions directly to management

For the Community

• Vote on Answers - upvote helpful responses, downvote unhelpful ones
• Reputation System - earn points for asking, answering, and helping (Newcomer > Contributor > Helper > Expert > Guardian > Legend)
• Stories - share longer health journey articles
• Workplace Support - anonymous discussion boards for workplace wellbeing

AI-Powered

• Gemini AI Responses - every question gets an instant AI answer using Google Gemini 2.5 Flash
• Practical Advice First - AI trained to give actionable steps before medical referrals
• NHS-Aligned - uses UK terminology (GP, A&E, paracetamol) and references UK law

Admin Dashboard

• Full Control - delete questions/answers, manage users, promote/demote admins
• Hazard Report Management - acknowledge and resolve workplace hazard reports
• Site Statistics - questions, answers, users, AI answers at a glance
• Browser Push Notifications - real-time alerts when questions get answered

Tech Stack

Layer	Technology
Backend	Python, Flask
Database	PostgreSQL (production), SQLite (development)
ORM	SQLAlchemy with Flask-Migrate
Auth	Flask-Login, Google OAuth 2.0
AI	Google Gemini 2.5 Flash API
Frontend	Jinja2, Vanilla JavaScript, CSS
Voice	Web Speech API (browser-native)
Translation	Google Translate Widget
Health Data	NHS API Integration
Deployment	Docker, Render

Quick Start

Prerequisites

• Python 3.11+
• PostgreSQL (or SQLite for dev)

Setup

# Clone
git clone https://github.com/Pawansingh3889/Hackathon-mediask.git
cd Hackathon-mediask

# Install dependencies
pip install -r requirements.txt

# Set environment variables
export SECRET_KEY=your-secret-key
export DATABASE_URL=sqlite:///mediask.db
export GEMINI_API_KEY=your-gemini-key        # optional, for AI answers
export GOOGLE_CLIENT_ID=your-oauth-id        # optional, for Google login
export GOOGLE_CLIENT_SECRET=your-oauth-secret # optional, for Google login

# Initialize database and seed data
python -c "from app import create_app, db; app = create_app(); app.app_context().push(); db.create_all()"
python scripts/seed_categories.py
python scripts/seed_qa.py
python scripts/seed_workers.py

# Run
python run.py

Open http://localhost:5000

Docker

docker build -t mediask .
docker run -p 5000:5000 -e DATABASE_URL=sqlite:///mediask.db mediask

Project Structure

Hackathon-mediask/
|-- app/
|   |-- __init__.py          # Flask app factory
|   |-- models.py            # SQLAlchemy models (User, Question, Answer, Vote, etc.)
|   |-- ai_responder.py      # Gemini AI auto-responder
|   |-- config.py            # Configuration
|   |-- auth/                # Authentication (login, register, Google OAuth)
|   |-- main/                # Homepage, search, admin dashboard, NHS API
|   |-- questions/           # Q&A (ask, detail, vote, delete)
|   |-- users/               # Profiles, notifications, reputation
|   |-- workplace/           # Workplace wellbeing boards
|   |-- stories/             # Health journey stories
|   |-- templates/           # Jinja2 HTML templates
|   |-- static/              # CSS, JavaScript, images
|-- scripts/
|   |-- seed_categories.py   # 17 health categories
|   |-- seed_qa.py           # 40+ health Q&A with AI answers
|   |-- seed_workers.py      # 30+ factory worker Q&A with AI answers
|   |-- backfill_ai_answers.py # Add AI answers to existing questions
|-- Dockerfile
|-- requirements.txt
|-- run.py

Data Sources

• NHS UK - Official health guidance (nhs.uk)
• GOV.UK / HSE - Workplace health & safety regulations
• WHO - World Health Organization recommendations
• CDC - US Centers for Disease Control data
• NHS API - Live condition search

Health Categories

Mental Health, Cancer, Cardiovascular, Surgery, Blood Disorders, Eye & Vision, ENT, Infections, Skin Conditions, Stroke, Congenital Disorders, Immune System, Metabolic & Endocrine, Allergies, Memory & Neurology, Relationships & Emotions, General Health, Workers Health

Author

Pawan Singh Kapkoti

• MSc Data Analytics, Aston University
• GitHub: @Pawansingh3889
• Location: Hull, UK

License

This project was built for a hackathon. All health information is sourced from public NHS and GOV.UK resources.