.png){kind=icon}
.png){kind=icon}
Return
ScribSAM: A robust scribble-supervised framework for spatiotemporal segmentation of breast lesions in ultrasound videos
.png){kind=icon}
PRE
AI
L
Q
Y
Q
DOI：10.1016/j.compmedimag.2026.102790.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Pioneers scribble-supervised segmentation in ultrasound videos by providing high-quality scribble annotations on existing datasets. • Proposes ScribSAM, a novel MedSAM-based framework that integrates flow-guided propagation and spatiotemporal fusion for sparse-label learning. • Achieves competitive segmentation performance using only 4% of annotation pixels, significantly reducing labeling costs while maintaining robust clinical accuracy.
Journal
IF:
4.9
Papers:
2.4K
Citations:
5.0K
Organization
Cited Papers