This project aims to fine-tune Vision Language Models (VLMs) on doctor handwriting to improve the readability and interpretation of medical prescriptions and notes. We're creating synthetic data to enhance model's performance and make it more accessible through the Decipher Doctor platform.
- Fine-tuning of Qwen2-VL for doctor handwriting recognition
- Synthetic data generation for improved model performance
- Open-source initiative to enhance medical communication
We're actively looking for compute resources to accelerate our synthetic data generation and model fine-tuning. If you can provide GPU time or other computational resources, please contact us.
Alexander Al-Feghali - alexander.al-feghali@mail.mcgill.ca
Website: Decipher.Doctor
Blog: Doctor Handwriting Reader
- Qwen2-VL
- Hugging Face Transformers
- OpenAI for inspiration and research in the field of AI
- Python 3.7+
- PyTorch
- Transformers library
- Pillow
- Matplotlib
- NumPy
- Clone the repository:
git clone https://github.com/yourusername/doctor-handwriting-reader.git - Install the required packages:
pip install -r requirements.txt
We welcome contributions to improve the project! If you're interested in helping, please:
- Fork the repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Distributed under the MIT License. See LICENSE for more information.