/OCR_EFR

QuickScanEFR provides a dedicated pipeline for extracting textual content from PFT (Pulmonary Function Testing) scan reports saved in PDF format at Foch Hospital database.

Primary LanguageJupyter NotebookMIT LicenseMIT

QuickScanEFR: OCR for Pulmonary Function Testing

QuickScanEFR provides a dedicated pipeline for extracting textual content from PFT (Pulmonary Function Testing) scan reports saved in PDF format at Foch Hospital database. Optimized specifically for PFT reports, it ensures precision in data extraction.

🚀 Features

  • PDF to Image Conversion:

    • Transform EFR PDF reports with embedded images or complex layouts into high-resolution PNG images.

  • Image Preprocessing:

    • Improve image quality from EFR reports for better OCR results.

  • OCR Processing:

    • Utilize Tesseract (OCRobot) and Camelot (TextMachina) integrated through a plumber pipeline.
    • Extract and structure automatically the data.

⚙️ Optimization

  • Employs parallel processing for enhanced performance.
  • Swift processing of multiple EFR reports, especially beneficial for multi-core systems.

🔮 Future Enhancements

  • Support for multi-page EFR PDF reports.
  • Boost OCR accuracy with further fine-tuning.
  • Intuitive GUI for a user-friendly experience.

🤝 Contributing

  • Interested in making a difference? Fork the repository and submit a pull request!

📜 License

  • QuickScanEFR is proudly licensed under the MIT License.