UI: https://onevisionary.github.io/blood_report_analyser.github.io/

Blood Report Summarizer is a Flask-based web application that uses AI-powered document parsing and summarization to analyze blood test PDF reports and generate clear, structured, and easy-to-understand summaries.

The application extracts text and layout information from medical PDFs and leverages large language models to help users better understand their lab results through a simple, modern web interface.

• PDF Upload Upload blood test reports using drag & drop or file browser

• Layout-Aware Text Extraction Extracts structured text and images from PDFs using Baidu Layout Parsing API Handles tables, headings, and complex medical report layouts

• AI-Powered Summarization Generates concise and structured summaries using OpenRouter AI Highlights key findings and abnormal values

• User-Friendly Interface Clean, responsive UI with progress tracking Displays extracted pages, raw text, and AI-generated summary

• Medical Disclaimer Clearly states that the output is AI-generated and not a replacement for medical professionals

• Backend: Flask (Python)
• PDF Processing: Baidu Layout Parsing API (ERNIE + PaddleOCR)
• AI Summarization: OpenRouter API
• Frontend: HTML, CSS, JavaScript
• Environment Management: Python virtual environment (venv)

• Python 3.8 or higher

• API keys for:

  • Baidu Layout Parsing API
  • OpenRouter API

git clone <repository-url>
cd blood-report-summarizer

python -m venv venv

Activate the environment:

Windows

venv\Scripts\activate

macOS / Linux

source venv/bin/activate

pip install -r requirements.txt

Create a .env file in the project root directory and add the following:

BAIDU_TOKEN=your_baidu_api_token_here
OPEN_ROUTER=your_openrouter_api_key_here

1. Visit Baidu AI Studio
2. Apply for Layout Parsing API access
3. Generate your API token
4. Add the token to the .env file

1. Visit OpenRouter
2. Sign up or log in
3. Generate an API key
4. Add the key to the .env file

The free tier is sufficient for testing and demos.

python app.py

The application will be available at:

http://localhost:5000

1. Upload a blood test report PDF using drag & drop or file selection

2. Click the Process Report button

3. View results:

   • Extracted report pages as images
   • Raw extracted text
   • AI-generated blood report summary

• Multi-language support
• Downloadable summary reports (PDF)
• Visual highlighting of abnormal values
• User authentication and report history

This project is intended for educational and demonstration purposes.