MediSimply - AI Medical Report Explainer

Inspiration

Medical reports are written for doctors, not patients. Lab results, X-rays, MRI, CT, and ultrasound reports are full of technical terms that leave most people confused and anxious. We saw this gap clearly and wanted to build a tool that takes those complex reports and turns them into simple, clear explanations anyone can understand — without needing a medical degree. That idea became MediSimply.

What I built

MediSimply is an AI-powered web app that helps patients understand their medical reports in plain language. Users can:

• Upload a report (PDF or image) or paste the text
• Get a clear, structured explanation of the findings
• See medical terms explained in simple words
• Choose a simpler reading level if needed Every response includes a strong disclaimer that the tool is only for educational purposes and is not a substitute for a doctor’s advice.

How I built it

I designed MediSimply as an end-to-end system:

1. Document handling – Users upload PDFs or images of reports. We used OCR to extract text from scanned documents and photos.
2. AI explanation engine – The extracted text is sent to a large language model with carefully designed prompts. The prompts instruct the model to:
   • Summarize the key findings
   • Explain medical terms in everyday language
   • Avoid giving any diagnosis or treatment advice
3. Clean output – Results are shown in a simple, readable format with clear sections so users can quickly understand what the report is saying.
4. Safety layer – We added explicit disclaimers and prompt constraints so the AI never crosses into medical advice.

The goal was to make the experience fast, trustworthy, and easy to use.

Challenges I faced

• Keeping explanations accurate but simple – It was hard to simplify medical language without losing important meaning or introducing mistakes. We spent a lot of time refining the prompts.
• OCR quality – Many reports are scanned or photographed poorly. Extracting clean text from messy documents was a major challenge.
• Staying safe – We had to carefully control the AI so it explains the report without diagnosing or advising. Balancing helpfulness and safety took several iterations.
• Privacy concerns – Medical data is sensitive. We designed the flow to process reports carefully and avoid unnecessary storage of personal health information.
• Time pressure – Building a usable, safe product within the challenge timeline forced us to focus on the core experience and leave some advanced features for later.

What we learned

We learned how powerful AI can be when it is guided carefully, especially in sensitive domains like healthcare. Good prompting, clear disclaimers, and a simple user experience matter more than adding every possible feature. We also realized how important it is to respect the limits of AI — it can explain, but it should never replace a doctor.

MediSimply is our attempt to make medical reports a little less scary and a lot more understandable.ner

Built With

• easyocr
• github
• groq
• huggingfacetransformers
• llm
• naturallanguageprocessing
• ocr
• pdfplumber
• python
• streamlitcloud