marwanaljijakli/ocr-rag-api repository overview

FastAPI-based OCR + RAG microservice that extracts biomarkers from lab report PDFs/images. Ships with Tesseract, OpenCV, FAISS, and OpenAI integration—all prebundled for instant deployments.

⁠OCR RAG API

A production-ready FastAPI service that converts lab report PDFs and medical images into structured biomarker data using OCR and retrieval-augmented generation (RAG).

⁠Key features

• Tesseract + OpenCV preprocessing for reliable OCR on complex lab reports
• FAISS vector index backed by OpenAI embeddings for biomarker retrieval
• REST endpoints for document upload, biomarker queries, and health checks
• Ready-to-run Docker image with all native dependencies installed

⁠Getting started

docker run --rm -p 8001:8001 \
  -e OPENAI_API_KEY=sk-your-key \
  marwanaljijakli/ocr-rag-api:latest