This repository will be a summary and outlook on all our open, medical, AI advancements. See the medical-general channel in the medical category on our discord server for more information!
- Fine-tune clip for pathology images [1]
- Replicate PLOS paper with current image generation techniques
- [1] image-text pairs of pathology images (~10k) Go to dataset
- TCGA Slide-caption pairs (~3K) with captions scraped from pathology reports Go to dataset Scraping methodology
- Fine-tuning existing CLIP models