Driven by textual knowledge: A Text-View Enhanced Knowledge Transfer Network for lung infection region segmentation

Lexin Fang, Xuemei Li, Yunyang Xu, Fan Zhang, Caiming Zhang

Abstract: Highlights•TVE-Net leverages lesion location info from medical reports for segmentation.•Probabilistic function maps text descriptions to adaptive probability maps.•Self-supervised constraint boosts features and semi-supervised performance.•The multi-stage knowledge transfer module enriches image features with text.