Medi-Genie

Inspiration

Medical reports are often filled with complex medical jargon, making it difficult for non-medical users to understand their health status.
Inspired by the need to make healthcare more accessible and transparent, we built an AI-powered tool that simplifies medical reports, making them easy to read and interpret.

What it does

βœ… Securely log in using Auth0 (Google, Email, Yahoo, Facebook, Twitter)
βœ… Upload medical reports in PDF, Image, or Text formats
βœ… Utilize GenAI to extract key insights and explain medical terms in simple language
βœ… Highlight abnormal vs. normal values in test reports
βœ… Offer personalized health suggestions based on report data
βœ… Track health trends over time with historical report storage
βœ… Generate reports in multiple languages, allowing accessibility for diverse users

How we built it

πŸš€ Tech Stack Used:

  • Frontend: React.js (for user interaction with uploading file and report)
  • Backend: Python (Flask)
  • Database: MongoDB (stores reports & analysis)
  • AI and Textract: for text analysis & simplification
  • Authentication: Auth0 (secure login via Google, Email, Yahoo, Facebook, or Twitter)
  • Storage: Local storage for file uploads

Challenges we ran into

πŸ”Ή Struggled with poor text extraction due to low-quality scans and complex medical terminology.
πŸ”Ή Difficulty extracting text from PDFs with embedded images or multi-column layouts.
πŸ”Ή Needed to prevent malicious file uploads and manage large file sizes.

Accomplishments that we're proud of

πŸ† Implemented secure authentication & role-based access control (RBAC) using Auth0 to manage user access and permissions seamlessly.
πŸ† Leveraged MongoDB Atlas for real-time data storage, fast queries with indexing, and secure database access using IP whitelisting.
πŸ† Integrated OpenAI’s GenAI for text summarization, intelligent recommendations, and conversational AI to enhance medical report analysis.
πŸ† Built a seamless React-Frontend and Flask-Backend communication for a smooth user experience.
πŸ† Enabled multi-format report uploads with precise text extraction.

What we learned

πŸ“Œ The importance of natural language processing (NLP) in simplifying medical jargon
πŸ“Œ Handling data security and privacy in medical applications
πŸ“Œ Optimizing AI models for better accuracy in text extraction and analysis
πŸ“Œ Enhancing user experience for better engagement and accessibility

What's next for MediGenie

🎯 Voice-Based Summarization: Provide AI-generated voice summaries for reports
🎯 More AI-powered Insights: Expand to detecting potential health risks and anomalies
🎯 Mobile App Version: Build an iOS & Android app for on-the-go accessibility
🎯 Integration with Wearables: Sync data from smartwatches for real-time health monitoring

Built With

Share this project:

Updates