A full-stack healthcare platform that connects patients with medical specialists using AI-powered matching, real government clinician data, and interactive medical visualization tools.

• Doctor Search & Discovery — Search 2.8M+ real clinicians from CMS government data by location, specialty, or symptoms
• AI-Powered Symptom Matching — Google Gemini AI analyzes symptoms and recommends appropriate medical specialties with confidence scoring
• Precision Consult Match — Calculates doctor-patient compatibility scores, predicts consultation fees, and supports appointment scheduling
• Interactive Map — View nearby doctors on an OpenStreetMap-powered map with distance and travel time calculations
• 3D Anatomy Viewer — Explore interactive 3D medical visualizations tied to specialties and conditions
• Clinical Nexus Academy — AI-driven medical education platform with condition explanations and specialist information
• AI Chat Assistant — Conversational medical guidance with patient info collection, image upload, and diagnostic suggestions
• User Accounts — Authentication, profile management, avatar uploads, and theme customization via Supabase
• Emergency Features — Emergency location sharing and contact alerts

• Node.js (v18 or later recommended)
• npm
• Git LFS (the CMS clinician CSV is tracked with LFS)

git clone https://github.com/koushikvasa/mediconnect.git
cd mediconnect

Frontend — create frontend/.env:

VITE_SUPABASE_URL=<your-supabase-url>
VITE_SUPABASE_ANON_KEY=<your-supabase-anon-key>
VITE_GEMINI_API_KEY=<your-gemini-api-key>

Backend — create backend/.env:

GEMINI_API_KEY=<your-gemini-api-key>
PORT=3001

# Backend
cd backend
npm install

# Frontend
cd ../frontend
npm install

cd backend
npm run import-data          # Import CMS CSV into SQLite
npm run generate-addresses   # Generate addresses for doctors

# Start the backend (port 3001)
cd backend
npm run dev

# In a separate terminal, start the frontend (port 5173)
cd frontend
npm run dev

Or run both together from the frontend directory:

cd frontend
npm run dev:with-backend

Open http://localhost:5173 in your browser.

mediconnect/
├── backend/
│   ├── controllers/          # Route handlers
│   ├── routes/               # API endpoint definitions
│   ├── services/             # Business logic (CSV processing, AI matching)
│   ├── scripts/              # Data import & address generation utilities
│   ├── data/                 # SQLite DB & CMS CSV (Git LFS)
│   └── server.js             # Express entry point
├── frontend/
│   ├── src/
│   │   ├── components/       # React components (map, 3D viewer, academy, etc.)
│   │   ├── services/         # API clients & AI service integrations
│   │   ├── hooks/            # Custom React hooks
│   │   ├── utils/            # Constants and helpers
│   │   ├── assets/           # Static images
│   │   ├── App.jsx           # Main application component
│   │   └── main.jsx          # React entry point
│   ├── server/
│   │   └── proxy-server.js   # Dev proxy server
│   └── vite.config.js        # Vite configuration
└── README.md

GET /api/doctors/search?location=<city>&specialty=<spec>&symptoms=<symptoms>&age=<age>&limit=20

POST /api/specialty/recommend
{
  "symptoms": "chest pain and shortness of breath",
  "age": 45,
  "gender": "male"
}

This project is for educational and demonstration purposes.