MedPrompt AI

Inspiration

Healthcare in many parts of the world, especially Africa, is often hard to access, slow, or expensive. I wanted to create a solution that allows anyone with a smartphone to quickly check symptoms, locate nearby help, and act fast in emergencies — without needing to understand medical jargon or face language barriers. MedPrompt AI was built to bring fast, AI-powered health support to people who need it most.

What it does

MedPrompt AI is an AI-powered health assistant that allows users to: Input symptoms and receive AI-generated health insights. Chat with a doctor-like assistant for advice. Use voice input to describe symptoms. Find nearby hospitals, pharmacies, and medical services. Trigger emergency protocols with one click (e.g., call 911 or emergency contact). Store health history securely using blockchain. Support multiple languages with Lingo AI. Toggle between dark and light modes for accessibility.

How we built it

Frontend: Built using React + Tailwind CSS, with support for mobile responsiveness and UI animations. Backend & Auth: Powered by Supabase for authentication and database. AI: Integrated Gemini AI for medical symptom analysis and responses. Voice Support: Used ElevenLabs for voice generation and feedback. Translation: Powered by Lingo to handle multilingual requests. Blockchain: Used a smart contract to store patient health records securely. Maps & Emergency Tools: Utilized Google Maps API and tel: protocol for emergency actions on mobile.

Challenges we ran into

Phone-based voice synthesis had unexpected behavior (e.g., repeating messages). Emergency call triggers are not fully supported on all browsers, so we had to add error messages and workarounds.

Accomplishments that we're proud of

Deployed a full-stack health assistant with AI, blockchain, and voice support in a short time. Designed a clean, user-friendly interface with both light and dark modes. Created a meaningful solution that could actually save lives in emergencies. Learned how to use tools like Supabase and ElevenLabs creatively.

What we learned

How to build and deploy a real-world AI health app with limited resources. The importance of considering legal and ethical implications in digital health. Troubleshooting and adapting to mobile vs. desktop behavior differences. How to blend AI, UX, and emergency tech to serve real needs.

What's next for MedPrompt AI

Partner with local healthcare providers and ambulance services. Expand the multilingual feature to include voice-to-voice translation. Optimize blockchain features for better health data portability.