RAGnosis is an AI- powered medical report analysis.

index page
index page
index page
user dashbaord
uplode medical report page
report result
health report tracking
chat baot

Inspiration

Healthcare reports are often complex and difficult for patients to understand. Many patients struggle with medical terminology, delayed consultations, and tracking their health progress over time. We were inspired to build RAGnosis to bridge the gap between raw medical data and patient understanding by using AI to simplify reports and provide intelligent insights for both patients and doctors.

What it does

RAGnosis is an AI-powered medical report analysis system that: Accepts uploaded medical reports (PDF/images) Uses a RAG (Retrieval Augmented Generation) engine to retrieve relevant medical knowledge Generates simple, easy-to-understand summaries Visualizes patient health progress through graphs and charts Provides an AI chatbot to answer patient queries related to reports Helps doctors quickly review patient history and insights

How we built it

Frontend: React.js for an interactive and responsive UI Backend: Python Flask for API handling and model integration AI Models: Hugging Face transformer models (BART, BERT) fine-tuned on medical datasets RAG Engine: Groq + Mixtral based retrieval and generation framework Database: MongoDB knowledge base for storing medical information Visualization: Pandas for structured data extraction and Matplotlib for progress graphs Chatbot: Transformer-based conversational model for report Q&A

Challenges we ran into

Handling different medical report formats (PDF, scanned images) Ensuring medical summary accuracy and avoiding hallucination Extracting structured data from unstructured report Optimizing RAG retrieval speed for real-time response Designing a UI simple enough for non-technical patients

Accomplishments that we're proud of

Successfully implemented a working RAG-based medical summarization system Built a chatbot capable of answering report-related questions Generated meaningful visual progress tracking graphs Created a clean and user-friendly interface Demonstrated real-world healthcare impact and research potential

What we learned

Practical implementation of Retrieval Augmented Generation (RAG) Fine-tuning and using transformer models in healthcare domain Data preprocessing and structured extraction from medical reports Integration of AI models with full-stack web applications Importance of explainable AI in healthcare

What's next for RAGnosis is an AI- powered medical report analysis.

Integration with EHR (Electronic Health Record) systems Expansion of the medical knowledge base for better retrieval accuracy Support for multi-language medical summaries Advanced NLP for disease prediction and risk alerts Mobile app deployment for accessibility Fletcher compliance and security enhancements for healthcare data