An AI-powered approach transforms routine pathology slides into dynamic maps of tumor evolution, enabling large-scale analysis of cancer progression without the need for costly single-cell sequencing.
- Model training and testing
- How to extract image patch features across the whole slide pathology images for the prediction of tumor grades
- How to calculate pseudotime, quantify tumor progression and downstream analysis of association with patient survivals
- python 3
- pandas==2.2.1
- torch==2.2.1+cu118
- torchvision==0.17.1+cu118
- transformers==4.38.2
- opencv-python==4.9.0.80
- openslide-python==1.3.1
- lifelines==0.28.0
- scikit-learn==1.4.1.post1
- scipy==1.12.0
- scanpy==1.9.8
- matplotlib==3.8.3
- seaborn==0.13.2
- anndata==0.10.6