MediLens

Understand what medicine you're taking in. No more guesswork. No more fake meds

Comment

🩺 Inspiration

A few months ago, my aunt (who lives in a rural town) was prescribed some medication after a visit to a local clinic. She didn’t fully understand what it was for, but she trusted the doctor.

Days later, she developed severe side effects - turns out, the medicine was a cheap counterfeit, and also dangerous for someone with diabetes like her.

We realized something painfully common in my country:

  • People take medicine they don’t understand.

  • Even if someone looked up the information of the medicine online, they are not likely to understand the complex medical terms. 

       Eg: When looking up Rosuvastin medicine, google says it "treats Hyperlipidemia and Primary Dysbetalipoproteinemia".

  • Fake drugs circulate, and patients pay the price.

MediLens was born out of that pain — and the hope that we can give people the power to understand and protect their own health, using technology.

💊 What it does

MediLens is a web app that helps users instantly understand any medicine and verify if it might be fake.

With just a scan or a typed-in name, MediLens provides:

  • ✅ What the medicine treats (in both medical terms and simple everyday language)
  • ⚠️ Common side effects
  • 🚫 Safety warnings (e.g., if unsafe for diabetics, pregnant women,)
  • 🔍 An Authenticity Checker to flag suspicious or counterfeit drugs

It's like giving every family their own trusted pharmacist - in their pocket.

🛠️ How we built it

  • Frontend: React + Tailwind CSS
  • AI integration: OpenAI GPT-4 API trained on medical data generates trusted medical summaries
  • Backend: Used SupaBase to handle Edge Functions
  • Fake Medicine Detection: Built a basic pattern-based checker to flag common counterfeit risks
  • Accessibility: Explanations in simple everyday terms for those who dont understand complex medical terms.
  • Data Structure: Follows a structured JSON schema to ensure clean UI rendering

We focused on shipping something real, fast, and functional — without compromising empathy.

🚧 Challenges we ran into

  • Ensuring the AI output was consistent and in strict JSON format for rendering
  • Translating medical knowledge into accurate but general population-friendly language
  • Adding OCR support with limited API access during the hackathon
  • Designing an intuitive UI for users who may not be tech-savvy or literate
  • Avoiding hallucinated or unsafe AI-generated medical info — we had to carefully design prompts

🏆 Accomplishments that we're proud of

  • 🧠 Built a working MVP that blends AI and health safety
  • 🩺 Delivered a project with real-world impact potential in rural and urban parts of the world
  • 🎯 Created a simple UI that can be used by both a medical college student and a rural grandmother
  • 💚 Made a tool that could save lives by preventing fake or wrongly used medicine

📚 What we learned

  • Simplicity wins ; a clear UI beats a flashy one when your users are in real need
  • Empathy in design is not optional in healthcare - it’s core functionality
  • There’s still huge untapped space for AI in preventive healthcare and health literacy in South Asia and beyond.

🚀 What's next for MediLens

  • ⚕️ Partner with real pharmacists to expand our medicine database and flag high-risk brands
  • 🤝 Add multi-language translation support for non-english speakers
  • 🌍 Add offline functionality for remote villages with limited internet
  • 📲 Launch a mobile version with voice search and real-time barcode scanning
  • 🧪 Add a “symptom to medicine” recommender for common health issues

Our mission is clear:

Every person deserves to know what they’re putting into their body. With MediLens, we’re making that knowledge accessible, trustworthy, and empowering for everyone.

Built With

Share this project:

Updates