.png){kind=icon}
.png){kind=icon}
Return
MTGT: Multiscale Text Feature-Guided Transformer in medical image segmentation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.imavis.2025.105846.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel segmentation paradigm with strong domain generalization is proposed. • The segmentation performance of the model is effectively improved by text annotation. • A multi-scale subtractive module is introduced to exploit differences across layers. • A text-guided attention module is designed to enhance text–image feature fusion. • Simulated and clinical verification of text annotations is performed on BUSI.
Journal
IF:
4.2
Papers:
4.0K
Citations:
6.7K