RadEdit: Using diffusion models to edit images for stress-testing biomedical vision models; RAD-DINO: Studying whether pre-training on multi-modal biomedical data can actually be a hindrance compared to uni-modal methods.
Research Experience
Senior researcher at the Health Futures group in Microsoft Research Cambridge, involved in projects like RadEdit and RAD-DINO.
Education
PhD from Durham University, where he explored how generative imaging models can be efficiently scaled to high resolutions.
Background
Senior researcher, focusing on multi-modal medical imaging models, particularly in generative imaging models and efficient scaling to high resolutions.