Health Bridge AI

Chat your symptoms, get smart health insights and relief tips. Your data stays private, stored as NFTs in your wallet—giving you full control of your personal health records.

Comment

🌟 Inspiration

In many parts of the world, early access to healthcare is either limited or intimidating. Many individuals delay seeking care due to cost, lack of access, or fear of judgment. We wanted to build a solution that empowers people to better understand their health—privately, instantly, and securely. The idea behind Health Bridge AI was to create a conversational, accessible tool that combines intelligent symptom checking with cutting-edge privacy features powered by Web3.

🧠 What We Learned

Through this project, we explored the intersection of AI-driven healthcare and decentralized data ownership. We learned how to:

  • Use symptom classification models and prompt engineering to deliver useful health insights.
  • Work with blockchain APIs like Verbwire to mint secure, non-transferable NFTs representing health data.
  • Design a chatbot that balances clarity, empathy, and responsibility in sensitive conversations.

🔧 How We Built It

Health Bridge AI was built over 24 hours using the following technologies:

  • Frontend: streamlit just for functioning for a clean and responsive user interface.
  • Backend: Python with Flask for chatbot logic and API handling.
  • AI: Open-source LLMs to interpret symptoms and generate possible condition matches and advice.
  • Web3: Verbwire API to mint user-specific NFTs containing encrypted health summaries.
  • Storage: IPFS to host the health reports securely and ensure data permanence.

Users interact with the chatbot by entering their symptoms. The backend processes this data and returns potential conditions and self-care suggestions. Once a report is generated, it is minted as an NFT directly into the user’s connected wallet, ensuring that only they can access it.

🚧 Challenges We Faced

  • Designing a responsible AI system that gives useful suggestions without overstepping the limits of non-clinical guidance.
  • Integrating blockchain functionality in a way that felt natural and privacy-focused—not gimmicky.
  • Balancing speed and accuracy within the symptom classification flow, especially under tight development timelines.
  • Managing deployment and interoperability across multiple stacks (AI, frontend, backend, blockchain) within a 24-hour window.

Built With

Share this project:

Updates