FGML-DG: Feynman-Inspired Cognitive Science Paradigm for Cross-Domain Medical Image Segmentation
FGML-DG is a meta-learning framework inspired by cognitive science to improve medical image segmentation across different data types. It mimics human learning by using concept understanding, memory recall, and targeted retraining to better transfer knowledge from seen domains to unseen ones. This approach significantly outperforms existing methods on complex medical tasks.