Inspiration
Medical jargon can be terrifying and confusing for patients. When a doctor writes a summary full of complex clinical terms, patients often turn to search engines, leading to unnecessary anxiety. We built MediTerms AI to bridge the communication gap between healthcare professionals and patients, translating complex medical notes into clear, empathetic, and actionable everyday language.
What it does
MediTerms AI is a simple, user-friendly web application. A user can paste a medical report, clinical summary, or prescription note into the app. Powered by an AI language model, the application instantly breaks down the text into three clear sections:
- Plain Translation: A breakdown of the medical terms in simple English.
- Key Takeaways: Bullet points highlighting the most important pieces of information.
- Suggested Next Steps: Clear, actionable questions or actions the patient should discuss with their doctor. How we built it As a beginner-focused project, we chose a highly efficient and modern tech stack to ensure quick deployment and a smooth user experience. The frontend is built using standard web technologies for a clean interface, while the backend utilizes a lightweight framework to securely communicate with a generative AI API (like OpenAI or Anthropic) using a custom system prompt designed for medical translation.
Challenges we faced One of our primary challenges was prompt engineering—ensuring the AI remains strictly a text translator and explicitly states it is not a medical diagnostic tool. We worked hard to ensure the tone is empathetic, clear, and reassuring without misinterpreting the clinical data.
What we learned We learned how to connect a user interface to an AI API, the importance of UI/UX design when handling sensitive topics like health, and how to manage a software project from a blank page to a working prototype within a strict hackathon timeline.
What's next for MediTerms AI We plan to add optical character recognition (OCR) so users can simply snap a photo of a physical paper report, expand language localization so patients can translate jargon into their native languages, and implement automated text-to-speech features for accessibility.
