MedSimplify

MedSi An overview of how medical reports are uploaded, processed using AI, and transformed into simple-language summaries

Inspiration

Medical reports contain important information about a person's health, but understanding medical terminology and test results can be difficult for people without a medical background. We were inspired to build MedSimplify to bridge this gap and make medical information easier to understand. We believe that patients should be able to understand their reports better and prepare meaningful questions for their healthcare professionals.

What it does

MedSimplify is an AI-powered medical report understanding platform. Users can upload medical reports in PDF, JPG, JPEG, and PNG formats, with a maximum file size of 15 MB.

The platform generates a summary in simple English, helps users explore their reports through an AI chatbot, and generates questions they can discuss with their doctors.

Our goal is to go beyond summarization by encouraging responsible understanding of medical information. A key direction for the project is to improve reliability by preserving the original test values and reference ranges and highlighting information that needs verification rather than allowing AI to guess.

How we built it

We built MedSimplify as a web application with a React-based frontend and a Python FastAPI backend. The project separates the user interface, report-processing functionality, and AI chatbot components to make the system easier to maintain and extend.

The frontend provides the report-upload experience and displays the generated explanations. The backend handles application requests and connects the relevant processing components. The AI chatbot is designed to answer users' follow-up questions about their reports in accessible language.

Challenges we ran into

One of our main challenges was making AI-generated explanations useful while maintaining accuracy in a healthcare-related application. Medical reports contain technical terms, numerical values, units, and reference ranges that must not be misinterpreted.

We also faced challenges integrating the frontend, backend, and AI components into a single workflow, as well as working with AI service configuration and API access.

These challenges taught us that building an AI application involves more than generating text: it also requires careful integration, testing, and responsible handling of information.

Accomplishments that we're proud of

We are proud of developing MedSimplify as a practical solution to a problem that affects many people. We brought together report uploading, simple-language summaries, interactive question-answering, and doctor-question generation into one platform.

We are also proud of identifying reliability as an important part of the solution. Our vision is to help users understand what their reports say while being transparent about uncertainty and the need for professional medical guidance.

What we learned

We learned how frontend and backend components work together in a web application, how AI can support a real-world use case, and how important testing and integration are when developing a complete product.

We also learned that healthcare-related AI requires particular care: explanations should be understandable, extracted information should be checked, and AI outputs should not be treated as medical diagnoses.

What's next for MedSimplify

Our next steps are to strengthen report extraction and verification, improve the grounding of chatbot responses in uploaded reports, and make uncertain or potentially abnormal findings easier to review.

We also aim to explore comparisons between reports from different dates, support for regional languages, and stronger privacy protections.

Our long-term vision is to make medical information more accessible and help people have better-informed conversations with their healthcare professionals. MedSimplify is intended to support understanding, not replace professional medical advice.