ClariMed

Inspiration

Navigating healthcare information is often overwhelming. Many patients leave hospitals with discharge summaries, lab results, or clinical notes filled with jargon that is difficult to understand, which can lead to confusion, anxiety, and missed follow-ups. This communication gap inspired us to create ClariMed — a platform designed to translate complex medical language into clear, actionable information. Our goal was to empower patients and caregivers to understand their health data, make informed decisions, and feel confident about their next steps in care.

What it does

ClariMed is an AI-powered web platform that transforms complicated medical documents into simple, easy-to-understand summaries. It explains technical terminology in plain language, highlights critical red-flag symptoms that require attention, and provides a clear list of recommended next steps. The platform also includes a built-in glossary of key medical terms for deeper understanding, offers one-click translation into multiple languages to improve accessibility, and features natural-sounding voice narration for users who prefer auditory explanations or need additional accessibility support. By converting confusing reports into clear, structured guidance, ClariMed enables patients and caregivers to actively participate in their healthcare journeys.

How we built it

We built ClariMed using a modern and scalable tech stack designed for seamless performance and accessibility. The frontend is built with React.js and styled with Tailwind CSS to deliver a responsive, user-friendly interface. The backend uses Flask in Python to handle API requests, orchestrate workflows, and manage data processing. For text extraction from scanned or uploaded medical documents, we integrated Google Cloud Vision’s OCR capabilities, while Google’s Gemini API powers text summarization, simplification, translation, and context-aware glossary generation. Snowflake serves as the backend database, storing and retrieving relevant medical knowledge and definitions, and ElevenLabs provides natural-sounding text-to-speech capabilities for narration. All files are securely stored in Google Cloud Storage, ensuring scalability and privacy throughout the process.

Challenges we ran into

One of the biggest challenges we faced was maintaining medical accuracy while simplifying highly technical language into accessible explanations. Striking the right balance between clarity and precision required extensive prompt tuning and validation. Another major hurdle was handling context-sensitive glossary definitions, especially when dealing with abbreviations or medical terms that have multiple meanings. OCR accuracy also posed difficulties when processing handwritten notes or low-quality scans, forcing us to implement additional validation layers. Additionally, designing a secure and privacy-conscious data flow was crucial, as handling sensitive healthcare information demanded careful consideration of compliance and safety best practices.

Accomplishments that we're proud of

Despite these challenges, we are proud of building a complete end-to-end system — from uploading a medical document to generating a clear, actionable summary — within a short time frame. We successfully integrated multilingual support and text-to-speech functionality, significantly improving accessibility for diverse users. Most importantly, we created a solution that addresses a real-world healthcare challenge and has the potential to improve health literacy, patient engagement, and outcomes for people around the world.

What we learned

Throughout the development process, we learned the importance of crafting context-aware AI prompts to ensure accurate, trustworthy, and human-centered explanations. We also gained valuable experience integrating multiple cloud services into a cohesive, scalable pipeline. Perhaps the most impactful lesson was the importance of user empathy: understanding the real anxieties and challenges patients face when reading medical documents guided many of our design and product decisions, ensuring that ClariMed remains patient-first at its core.

What's next for ClariMed

Looking ahead, we plan to expand ClariMed into a mobile application to make it even more accessible for patients and caregivers on the go. We also aim to integrate with electronic health record (EHR) systems using FHIR and HL7 standards, enabling automated retrieval of medical reports directly from healthcare providers. Another exciting step is the development of an interactive chatbot that can answer user questions in real time and provide deeper explanations. Finally, we envision creating an analytics dashboard for healthcare providers to track patient comprehension and engagement while enhancing our platform’s security and compliance to meet HIPAA and GDPR standards.