A GenAI-powered medical assistant that reduces misdiagnosis by combining patient data with PubMed + medical guidelines through RAG (Retrieval Augmented Generation).
✅ Secure login / authentication
✅ Patient symptom input form
✅ Medical history file upload (PDF/TXT)
✅ RAG search on PubMed + clinical guidelines
✅ Ranked diagnoses with confidence scores
✅ Recommended short-term & long-term treatments
✅ Interactive charts & visualizations
✅ Exportable PDF report
✅ Admin dashboard for case reviews
✅ Dark mode + mobile-first responsive UI
✅ Privacy/consent notice & disclaimers
- Frontend: Next.js 14, Tailwind CSS, Framer Motion
- Backend: Node.js, Next.js API Routes
- ML / GenAI: OpenAI API, FAISS for vector search, PubMed integration
- Database: PostgreSQL / Supabase (free tier)
- Deployment: Vercel
- Version Control: Git & GitHub
git clone https://github.com/your-username/your-repo.git
cd your-repo
2️⃣ Install dependencies
bash
Copy code
npm install
3️⃣ Run locally
bash
Copy code
npm run dev
Visit 👉 http://localhost:3000
4️⃣ Build for production
bash
Copy code
npm run build
npm start
📊 Example Output
json
Copy code
{
"diagnosis": "Type 2 Diabetes",
"confidence": "92%",
"treatment": {
"short_term": ["Control blood sugar with diet", "Start metformin if advised"],
"long_term": ["Weight management", "Exercise regularly", "Annual eye exams"]
},
"sources": [
"PubMed: https://pubmed.ncbi.nlm.nih.gov/xxxxxxx",
"American Diabetes Association Guidelines"
]
}
🎥 Demo / Screenshots
<p align="center"> <img src="https://media.giphy.com/media/xT9Igp0kGqF1zC3S3m/giphy.gif" width="700" alt="Demo GIF"> </p>
⚡ Deployment
Default: Deploy easily on Vercel
Or run in Docker:
bash
Copy code
docker build -t medrag .
docker run -p 3000:3000 medrag
📜 License
This project is licensed under the MIT License — see the LICENSE file for details.
⚠️ Disclaimer
⚠️ Not a substitute for professional medical advice.
This project is for educational and research purposes only.
Always consult a certified doctor for medical decisions.
<p align="center">Made with ❤️ using Next.js + RAG</p> ```